I received the M.S. degree from the School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, India, in 2009. After receiving the M.S. degree, I worked as a software engineer for a year, New Delhi, India. Afterward, I pursued my Doctoral degree with the Department of Computer Science, National Chiao Tung University (NCTU) and was awarded my Ph.D. degree in 2015. After my Ph.D. degree, I joined as Postdoctoral fellow (PDF) at NCTU. After successful completion of PDF, I joined as a principle engineer (R&D) with Taiwan Semiconductor Manufacturing Company, Hsinchu, Taiwan and worked till Feb. 2017. I started my academic career as Lecturer with University of Technology Sydney in March 2017 and became a core member of Centre for Artificial Intelligence(CAI). I have made substantial contributions to the field of machine learning and image processing, with 41 research articles (20 published/accepted) in top journals including IEEE Transactions and SCI/SCEI Journals and 21 conference papers.
Serve as an Associate Editor:
- International Journal of Business Intelligence and Data Mining
- International Journal of Intelligent Autonomous Systems
Serve as a reviewer for many International Journals and conferences:
- IEEE Transactions on Systems, Man and Cybernetics: Systems
- IEEE Transactions on Cybernetics
- IEEE Transactions on Industrial Electronics
- International Journal of Fuzzy Systems
- Neural Computing and Applications
- IET Wireless Sensor Systems
- International Journal of Approximate Reasoning
Serve as a Program/Session chair/co-chair at conference:
- Panel chair at IEEE-Fuzz 2017 for Brain Machine Interface, Naples, Italy
- International Advisor Committee, ICACCT 2017, New Delhi, India
- Poster session chair at INNS-Big Data 2018, Bali, Indonesia
- Program co-chair at AusPDC 2018, Sydney, Australia
- Web chiar at IEEE SSCI 2018, Banglore, India
Can supervise: YES
Machine Learning, Computational Intelligence, Deep Learning, Data Analysis, Fuzzy Neural Networks, Brain-computer Interface, Image processing.
- Introduction to Data Analytics (31250)
- Fundamentals of Data Analytics (32130)
Cheng, EJ, Prasad, M, Yang, J, Khanna, P, Chen, BH, Tao, X, Young, KY & Lin, CT 2020, 'A fast fused part-based model with new deep feature for pedestrian detection and security monitoring', Measurement: Journal of the International Measurement Confederation, vol. 151.View/Download from: UTS OPUS or Publisher's site
© 2019 Elsevier Ltd In recent years, pedestrian detection based on computer vision has been widely used in intelligent transportation, security monitoring, assistance driving and other related applications. However, one of the remaining open challenges is that pedestrians are partially obscured and their posture changes. To address this problem, deformable part model (DPM) uses a mixture of part filters to capture variation in view point and appearance and achieves success for challenging datasets. Nevertheless, the expensive computation cost of DPM limits its ability in the real-time application. This study propose a fast fused part-based model (FFPM) for pedestrian detection to detect the pedestrians efficiently and accurately in the crowded environment. The first step of the proposed method trains six Adaboost classifiers with Haar-like feature for different body parts (e.g., head, shoulders, and knees) to build the response feature maps. These six response feature maps are combined with full-body model to produce spatial deep features. The second step of the proposed method uses the deep features as an input to support vector machine (SVM) to detect pedestrian. A variety of strategies is introduced in the proposed model, including part-based to full-body method, spatial filtering, and multi-ratios combination. Experiment results show that the proposed FFPM method improves the computation speed of DPM and maintains the performance in detection.
King, JT, Prasad, M, Tsai, T, Ming, YR & Lin, CT 2020, 'Influence of Time Pressure on Inhibitory Brain Control During Emergency Driving', IEEE Transactions on Systems Man and Cybernetics: Systems.View/Download from: UTS OPUS or Publisher's site
IEEE It is believed that failures of people's reaction to emergencies occurred during driving are closely related to the inhibitory mechanism of brain's operations. To investigate the role of this function in emergency driving, two virtual realistic driving conditions based on stop signal task were designed and time limitation was manipulated to increase the stress in one condition. Sixteen subjects with behavioral encephalography recordings were collected and analyzed. By comparing successful and unsuccessful stop trials with event-related spectral perturbation analysis, δ and θ band power increases in frontal and central areas are correlated with driving inhibitory control of the brain. Moreover, β and ɣ band power in frontal and central areas showed more increases upon stress condition. Time pressure in driving could adjust the operation of brain's inhibition control, to benefit the people's reactive ability upon emergency.
Liu, Z, Cao, J, Tan, Y, Xiao, Q & Prasad, M 2020, 'Planning Above the API Clouds Before Flying Above the Clouds: A Real-Time Personalized Air Travel Planning Approach', International Journal of Parallel Programming, vol. 48, no. 1, pp. 137-156.View/Download from: Publisher's site
© 2019, Springer Science+Business Media, LLC, part of Springer Nature. The rapid growth of the airline industry has resulted in the availability of a large number of flights, however this can also create a paralyzing problem. Flight information on all airlines across the world can be obtained via the Internet. Today, passengers trend to be interested in user personalized service. How to effectively find a passenger’s most preferred air travel plan, which might include multiple transfers from millions of possible choices with certain constraints, such as time and price, is a critical challenge. This paper presents an efficient air travel planning approach, which can find a number of air travel plans by invoking the APIs offered by airline companies. At the same time, these plans also best match the customer’s preference based on an analysis of historical orders. An algorithm to extract user preference features is introduced and heuristic rules to speed up the K path search process under constraints are presented. The experiment results show that the proposed model finds optimal air travel plans efficiently on a real-world dataset.
Prakash, S, Joshi, S, Bhatia, T, Sharma, S, Samadhiya, D, Shah, RR, Kaiwartya, O & Prasad, M 2020, 'Characteristic of enterprise collaboration system and its implementation issues in business management', International Journal of Business Intelligence and Data Mining, vol. 16, no. 1, pp. 49-65.View/Download from: Publisher's site
© 2020 Inderscience Enterprises Ltd. Collaboration is an extremely useful area for the most of the enterprise systems particularly within Web 2.0 and Enterprise 2.0. The collaboration provides help in enterprise collaboration system (ECS) to achieve the desired goal by unifying completed tasks of employees or people working on a similar or the same task. Thus, the collaboration systems have witnessed significant attention. The ECS provides consistent and off-the-shelf support to processes and managements within organisations. Management techniques of the ECS may be useful to a community which manages ECS systems for collaboration. In this context, this paper focuses on enterprise collaboration system and answers critical questions related to ECS including: 1) what does collaboration really means for an enterprise system; 2) how can the collaboration help to improve internal processes and management of the system; 3) how it is helpful to improve interactions with customers and partners?
Aliyu, A, El-Sayed, H, Abdullah, AH, Alam, I, Li, J & Prasad, M 2019, 'Video streaming in urban vehicular environments: Junction-aware multipath approach', Electronics (Switzerland), vol. 8, no. 11.View/Download from: UTS OPUS or Publisher's site
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. In multipath video streaming transmission, the selection of the best vehicle for video packet forwarding considering the junction area is a challenging task due to the several diversions in the junction area. The vehicles in the junction area change direction based on the different diversions, which lead to video packet drop. In the existing works, the explicit consideration of different positions in the junction areas has not been considered for forwarding vehicle selection. To address the aforementioned challenges, a Junction-Aware vehicle selection for Multipath Video Streaming (JA-MVS) scheme has been proposed. The JA-MVS scheme considers three different cases in the junction area including the vehicle after the junction, before the junction and inside the junction area, with an evaluation of the vehicle signal strength based on the signal to interference plus noise ratio (SINR), which is based on the multipath data forwarding concept using greedy-based geographic routing. The performance of the proposed scheme is evaluated based on the Packet Loss Ratio (PLR), Structural Similarity Index (SSIM) and End-to-End Delay (E2ED) metrics. The JA-MVS is compared against two baseline schemes, Junction-Based Multipath Source Routing (JMSR) and the Adaptive Multipath geographic routing for Video Transmission (AMVT), in urban Vehicular Ad-Hoc Networks (VANETs).
Gupta, D, Pratama, M, Ma, Z, Li, J & Prasad, M 2019, 'Financial time series forecasting using twin support vector regression.', PloS one, vol. 14, no. 3.View/Download from: UTS OPUS or Publisher's site
Financial time series forecasting is a crucial measure for improving and making more robust financial decisions throughout the world. Noisy data and non-stationarity information are the two key factors in financial time series prediction. This paper proposes twin support vector regression for financial time series prediction to deal with noisy data and nonstationary information. Various interesting financial time series datasets across a wide range of industries, such as information technology, the stock market, the banking sector, and the oil and petroleum sector, are used for numerical experiments. Further, to test the accuracy of the prediction of the time series, the root mean squared error and the standard deviation are computed, which clearly indicate the usefulness and applicability of the proposed method. The twin support vector regression is computationally faster than other standard support vector regression on the given 44 datasets.
Patel, OP, Bharill, N, Tiwari, A, Patel, V, Gupta, O, Cao, J, Li, J & Prasad, M 2019, 'Advanced quantum based neural network classifier and its application for objectionable web content filtering', IEEE Access, vol. 7, pp. 98069-98082.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. In this paper, an Advanced Quantum-based Neural Network Classifier (AQNN) is proposed. The proposed AQNN is used to form an objectionable Web content filtering system (OWF). The aim is to design a neural network with a few numbers of hidden layer neurons with the optimal connection weights and the threshold of neurons. The proposed algorithm uses the concept of quantum computing and genetic concept to evolve connection weights and the threshold of neurons. Quantum computing uses qubit as a probabilistic representation which is the smallest unit of information in the quantum computing concept. In this algorithm, a threshold boundary parameter is also introduced to find the optimal value of the threshold of neurons. The proposed algorithm forms neural network architecture which is used to form an objectionable Web content filtering system which detects objectionable Web request by the user. To judge the performance of the proposed AQNN, a total of 2000 (1000 objectionable + 1000 non-objectionable) Website's contents have been used. The results of AQNN are also compared with QNN-F and well-known classifiers as backpropagation, support vector machine (SVM), multilayer perceptron, decision tree algorithm, and artificial neural network. The results show that the AQNN as classifier performs better than existing classifiers. The performance of the proposed objectionable Web content filtering system (OWF) is also compared with well-known objectionable Web filtering software and existing models. It is found that the proposed OWF performs better than existing solutions in terms of filtering objectionable content.
Cheng, EJ, Chou, KP, Rajora, S, Jin, BH, Tanveer, M, Lin, CT, Young, KY, Lin, WC & Prasad, M 2019, 'Deep Sparse Representation Classifier for facial recognition and detection system', Pattern Recognition Letters, vol. 125, pp. 71-77.View/Download from: UTS OPUS or Publisher's site
© 2019 Elsevier B.V. This paper proposes a two-layer Convolutional Neural Network (CNN) to learn the high-level features which utilizes to the face identification via sparse representation. Feature extraction plays a vital role in real-world pattern recognition and classification tasks. The details description of the given input face image, significantly improve the performance of the facial recognition system. Sparse Representation Classifier (SRC) is a popular face classifier that sparsely represents the face image by a subset of training data, which is known as insensitive to the choice of feature space. The proposed method shows the performance improvement of SRC via a precisely selected feature exactor. The experimental results show that the proposed method outperforms other methods on given datasets.
Gupta, A, Agrawal, RK, Kirar, JS, Andreu-Perez, J, Ding, W-P, Lin, C-T & Prasad, M 2019, 'On the Utility of Power Spectral Techniques With Feature Selection Techniques for Effective Mental Task Classification in Noninvasive BCI', IEEE Transactions on Systems, Man, and Cybernetics: Systems, pp. 1-13.View/Download from: Publisher's site
Gupta, A, Agrawal, RK, Kirar, JS, Kaur, B, Ding, W, Lin, CT, Javier, AP & Prasad, M 2019, 'A hierarchical meta-model for multi-class mental task based brain-computer interfaces', Neurocomputing.View/Download from: UTS OPUS or Publisher's site
© 2019 In the last few years, many research works have been suggested on Brain-Computer Interface (BCI), which assists severely physically disabled persons to communicate directly with the help of electroencephalogram (EEG)signal, generated by the thought process of the brain. Thought generation inside the brain is a dynamic process, and plenty thoughts occur within a small time window. Thus, there is a need for a BCI device that can distinguish these various ideas simultaneously. In this research work, our previous binary-class mental task classification has been extended to the multi-class mental task problem. The present work proposed a novel feature construction scheme for multi mental task classification. In the proposed method, features are extracted in two phases. In the first step, the wavelet transform is used to decompose EEG signal. In the second phase, each feature component obtained is represented compactly using eight parameters (statistical and uncertainty measures). After that, a set of relevant and non-redundant features is selected using linear regression, a multivariate feature selection approach. Finally, optimal decision tree based support vector machine (ODT-SVM)classifier is used for multi mental task classification. The performance of the proposed method is evaluated on the publicly available dataset for 3-class, 4-class, and 5-class mental task classification. Experimental results are compared with existing methods, and it is observed that the proposed plan provides better classification accuracy in comparison to the existing methods for 3-class, 4-class, and 5-class mental task classification. The efficacy of the proposed method encourages that the proposed method may be helpful in developing BCI devices for multi-class classification.
Lin, CT, King, JT, Bharadwaj, P, Chen, CH, Gupta, A, Ding, W & Prasad, M 2019, 'EOG-Based Eye Movement Classification and Application on HCI Baseball Game', IEEE Access, vol. 7, pp. 96166-96176.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. Electrooculography (EOG) is considered as the most stable physiological signal in the development of human-computer interface (HCI) for detecting eye-movement variations. EOG signal classification has gained more traction in recent years to overcome physical inconvenience in paralyzed patients. In this paper, a robust classification technique, such as eight directional movements is investigated by introducing a concept of buffer along with a variation of the slope to avoid misclassification effects in EOG signals. Blinking detection becomes complicated when the magnitude of the signals are considered. Hence, a correction technique is introduced to avoid misclassification for oblique eye movements. Meanwhile, a case study has been considered to apply these correction techniques to HCI baseball game to learn eye-movements.
Ming, Y, Ding, W, Pelusi, D, Wu, D, Wang, YK, Prasad, M & Lin, CT 2019, 'Subject adaptation network for EEG data analysis', Applied Soft Computing Journal, vol. 84.View/Download from: UTS OPUS or Publisher's site
© 2019 Elsevier B.V. Biosignals tend to display manifest intra- and cross-subject variance, which generates numerous challenges for electroencephalograph (EEG) data analysis. For instance, in the context of classification, the discrepancy between EEG data can make the trained model generalising poorly for new test subjects. In this paper, a subject adaptation network (SAN) inspired by the generative adversarial network (GAN) to mitigate different variances is proposed for analysing EEG data. First the challenges faced by traditional approaches employed for EEG signal processing are emphasised. Then the problem is formulated from mathematical perspective to highlight the key points in resolving such discrepancies. Third, the motivation behind and design principle of the SAN are described in an intuitive manner to reflect its suitability for analysing EEG data. Then after depicting the overall architecture of the SAN, several experiments are used to justify the practicality and efficiency of using the proposed model from different perspectives. For instance, an EEG dataset captured during a stereotypical neurophysiological experiment called the VEP oddball task is utilised to demonstrate the performance improvement achieved by running the SAN.
Bharill, N, Patel, OP, Tiwari, A, Mu, L, Li, D-L, Mohanty, M, Kaiwartya, O & Prasad, M 2019, 'A Generalized Enhanced Quantum Fuzzy Approach for Efficient Data Clustering', IEEE ACCESS, vol. 7, pp. 50347-50361.View/Download from: UTS OPUS or Publisher's site
Bharill, N, Tiwari, A, Malviya, A, Patel, OP, Gupta, A, Puthal, D, Saxena, A & Prasad, M 2019, 'Fuzzy knowledge based performance analysis on big data', Neurocomputing.View/Download from: Publisher's site
© 2019 Elsevier B.V. Due to the various emerging technologies, an enormous amount of data, termed as Big Data, gets collected every day and can be of great use in various domains. Clustering algorithms that store the entire data into memory for analysis become unfeasible when the dataset is too large. Many clustering algorithms present in the literature deal with the analysis of huge amount of data. The paper discusses a new clustering approach called an Incremental Random Sampling with Iterative Optimization Fuzzy c-Means (IRSIO-FCM)algorithm. It is implemented on Apache Spark, a framework for Big Data processing. Sparks works really well for iterative algorithms by supporting in-memory computations, scalability, etc. IRSIO-FCM not only facilitates effective clustering of Big Data but also performs storage space optimization during clustering. To establish a fair comparison of IRSIO-FCM, we propose an incremental version of the Literal Fuzzy c-Means (LFCM)called ILFCM implemented in Apache Spark framework. The experimental results are analyzed in terms of time and space complexity, NMI, ARI, speedup, sizeup, and scaleup measures. The reported results show that IRSIO-FCM achieves a significant reduction in run-time in comparison with ILFCM.
Patel, OP, Bharill, N, Tiwari, A & Prasad, M 2019, 'A Novel Quantum-inspired Fuzzy Based Neural Network for Data Classification', IEEE Transactions on Emerging Topics in Computing.View/Download from: UTS OPUS or Publisher's site
IEEE The performance of the neural network (NN) depends on the various parameters such as structure, initial weight, number of hidden layer neurons, and learning rate. The improvement in classification performance of NN without changing its structure is a challenging issue. This paper proposes a novel learning model called Quantum-inspired Fuzzy Based Neural Network (Q-FNN) to solve two-class classification problems. In the proposed model, NN architecture is formed constructively by adding neurons in the hidden layer and learning is performed using the concept of Fuzzy c-Means (FCM) clustering, where the fuzziness parameter (m) is evolved using the quantum computing concept. The quantum computing concept provides a large search space for a selection of m, which helps in finding the optimal weights and also optimizes the network architecture. This paper also proposes a modified step activation function for the formation of hidden layer neurons, which handles the overlapping samples belong to different class regions. The performance of the proposed Q-FNN model is superior and competitive with the state-of-the-art methods in terms of accuracy, sensitivity, and specificity on 15 real-world benchmark datasets.
Patel, OP, Tiwari, A, Chaudhary, R, Nuthalapati, SV, Bharill, N, Prasad, M, Hussain, FK & Hussain, OK 2019, 'Enhanced quantum-based neural network learning and its application to signature verification', Soft Computing, vol. 23, no. 9, pp. 3067-3080.View/Download from: UTS OPUS or Publisher's site
© 2017 Springer-Verlag GmbH Germany, part of Springer Nature In this paper, an enhanced quantum-based neural network learning algorithm (EQNN-S) which constructs a neural network architecture using the quantum computing concept is proposed for signature verification. The quantum computing concept is used to decide the connection weights and threshold of neurons. A boundary threshold parameter is introduced to optimally determine the neuron threshold. This parameter uses min, max function to decide threshold, which assists efficient learning. A manually prepared signature dataset is used to test the performance of the proposed algorithm. To uniquely identify the signature, several novel features are selected such as the number of loops present in the signature, the boundary calculation, the number of vertical and horizontal dense patches, and the angle measurement. A total of 45 features are extracted from each signature. The performance of the proposed algorithm is evaluated by rigorous training and testing with these signatures using partitions of 60–40 and 70–30%, and a tenfold cross-validation. To compare the results derived from the proposed quantum neural network, the same dataset is tested on support vector machine, multilayer perceptron, back propagation neural network, and Naive Bayes. The performance of the proposed algorithm is found better when compared with the above methods, and the results verify the effectiveness of the proposed algorithm.
Chou, K-P, Prasad, M, Wu, D, Sharma, N, Li, D-L, Lin, Y-F, Blumenstein, M, Lin, W-C & Lin, C-T 2018, 'Robust Feature-Based Automated Multi-View Human Action Recognition System', IEEE ACCESS, vol. 6, pp. 15283-15296.View/Download from: UTS OPUS or Publisher's site
Ding, W, Lin, CT & Prasad, M 2018, 'Hierarchical co-evolutionary clustering tree-based rough feature game equilibrium selection and its application in neonatal cerebral cortex MRI', Expert Systems with Applications, vol. 101, pp. 243-257.View/Download from: UTS OPUS or Publisher's site
© 2018 Elsevier Ltd A wide variety of feature selection methods have been developed as promising solutions to find the classification pattern inside increasing applications. But the exploring efficient, flexible and robust feature selection method to handle the rising big data is still an exciting challenge. This paper presents a novel hierarchical co-evolutionary clustering tree-based rough feature game equilibrium selection algorithm (CTFGES). It aims to select out the high-quality feature subsets, which can enrich the research of feature selection and classification in the heterogeneous big data. Firstly, we construct a flexible hierarchical co-evolutionary clustering tree model to speed up the process of feature selection, which can effectively extract the features from the parent and children branches of four-layer co-evolutionary clustering tree. Secondly, we design a mixed co-evolutionary game equilibrium scheme with adaptive dynamics to guide parent and children branch subtrees to approach the optimal equilibrium regions, and enable their feature sets to converge stably to the Nash equilibrium. So both noisy heterogeneous features and non-identified redundant ones can be further eliminated. Finally, the extensive experiments on various big datasets are conducted to demonstrate the more excellent performance of CTFGES, in terms of accuracy, efficiency and robustness, compared with the representative feature selection algorithms. In addition, the proposed CTFGES algorithm has been successfully applied into the feature segmentation of large-scale neonatal cerebral cortex MRI with varying noise ratios and intensity non-uniformity levels. The results indicate that it can be adaptive to derive from the cortical folding surfaces and achieves the satisfying consistency with medical experts, which will be potential significance for successfully assessing the impact of aberrant brain growth on the neurodevelopment of neonatal cerebrum.
Ding, W, Lin, CT, Prasad, M, Cao, Z & Wang, JD 2018, 'A Layered-Coevolution-Based Attribute-Boosted Reduction Using Adaptive Quantum Behavior PSO and Its Consistent Segmentation for Neonates Brain Tissue', IEEE Transactions on Fuzzy Systems, vol. 26, no. 3.View/Download from: UTS OPUS or Publisher's site
IEEE The main challenge of attribute reduction in large data applications is to develop a new algorithm to deal with large, noisy, and uncertain large data linking multiple relevant data sources, structured or unstructured. This paper proposes a new and efficient layered-coevolution-based attribute-boosted reduction algorithm (LCQ-ABR*) using adaptive quantum behavior particle swarm optimization (PSO). First, the quantum rotation angle of an evolutionary particle is updated by a dynamic change of self-adapting step size. Second, a self-adaptive partitioning strategy is employed to group particles into different memeplexes, and the quantum-behavior mechanism with the particles & #x0027; states depicted by the wave function cooperates to achieve superior performance in their respective memeplexes. Third, a new layered co-evolutionary model with multi-agent interaction is constructed to decompose a complex attribute set, and it can self-adapt the attribute sizes among different layers and produce the reasonable decompositions by exploiting any interdependency among multiple relevant attribute subsets. Fourth, the decomposed attribute subsets are evolved to compute the positive region and discernibility matrix by using their best quantum particles, and the global optimal reduction set is induced successfully. Finally, extensive comparative experiments are provided to illustrate that LCQ-ABR* has better feasibility and effectiveness of attribute reduction on large-scale and uncertain dataset problems with complex noise, compared with representative algorithms. Moreover, LCQ-ABR* can be successfully applied in the consistent segmentation for neonatal brain 3D-MRI, and the consistent segmentation results further demonstrate its stronger applicability.
El-Sayed, H, Sankar, S, Prasad, M, Puthal, D, Gupta, A, Mohanty, M & Lin, CT 2018, 'Edge of Things: The Big Picture on the Integration of Edge, IoT and the Cloud in a Distributed Computing Environment', IEEE Access, vol. 6, pp. 1706-1717.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. A centralized infrastructure system carries out existing data analytics and decision-making processes from our current highly virtualized platform of wireless networks and the Internet of Things (IoT) applications. There is a high possibility that these existing methods will encounter more challenges and issues in relation to network dynamics, resulting in a high overhead in the network response time, leading to latency and traffic. In order to avoid these problems in the network and achieve an optimum level of resource utilization, a new paradigm called edge computing (EC) is proposed to pave the way for the evolution of new age applications and services. With the integration of EC, the processing capabilities are pushed to the edge of network de vices such as smart phones, sensor nodes, wearables, and on-board units, where data analytics and knowledge generation are performed which removes the necessity for a centralized system. Many IoT applications, such as smart cities, the smart grid, smart traffic lights, and smart vehicles, are rapidly upgrading their applications with EC, significantly improving response time as well as conserving network resources. Irrespective of the fact that EC shifts the workload from a centralized cloud to the edge, the analogy between EC and the cloud pertaining to factors such as resource management and computation optimization are still open to research studies. Hence, this paper aims to validate the efficiency and resourcefulness of EC. We extensively survey the edge systems and present a comparative study of cloud computing systems. After analyzing the different network properties in the system, the results show that EC systems perform better than cloud computing systems. Finally, the research challenges in implementing an EC system and future research directions are discussed.
Lin, CT, King, JT, Fan, JW, Appaji, A & Prasad, M 2018, 'The Influence of Acute Stress on Brain Dynamics during Task Switching Activities', IEEE Access, vol. 6, pp. 3249-3255.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. Task switching is a common method to investigate executive functions such as working memory and attention. This paper investigates the effect of acute stress on brain activity using task switching. Surprisingly few studies have been conducted in this area. There is behavioral and physiological evidence to indicate that acute stress makes the participants more tense which results in a better performance. In this current study, under stressful conditions, the participants gave quick responses with high accuracy. However, unexpected results were found in relation to salivary cortisol. Furthermore, the electroencephalogram results showed that acute stress was pronounced at the frontal and parietal midline cortex, especially on the theta, alpha, and gamma bands. One possible explanation for these results may be that the participants changed their strategy in relation to executive functions during stressful conditions by paying more attention which resulted in a higher working memory capacity which enhanced performance during the task switching.
Lin, CT, King, JT, Singh, AK, Prasad, M, Ma, Z, Lin, JW, Machado, AMC, Appaji, A & Gupta, A 2018, 'Voice Navigation Effects on Real-World Lane Change Driving Analysis using an Electroencephalogram', IEEE Access, vol. 6.View/Download from: UTS OPUS or Publisher's site
OAPA Improving the degree of assistance given by in-car navigation systems is an important issue for the safety of both drivers and passengers. There is a vast body of research that assesses the usability and interfaces of the existing navigation systems but very few investigations study the impact on the brain activity based on navigation-based driving. In this study, a real-world experiment is designed to acquire the electroencephalography (EEG) and in-car information to analyze the dynamic brain activity while the driver is performing the lane-changing task based on the auditory instructions from an in-car navigation system. The results show that auditory cues can influence the speed and increase the frontal EEG delta and beta power which is related to motor preparation and decision making during a lane change. However, there were no significant results on the alpha power. A better lane-change assessment can be obtained using specific vehicle information (lateral acceleration and heading angle) with EEG features for future naturalized driving study.
Lin, CT, Prasad, M, Chung, CH, Puthal, D, El-Sayed, H, Sankar, S, Wang, YK, Singh, J & Sangaiah, AK 2018, 'IoT-Based Wireless Polysomnography Intelligent System for Sleep Monitoring', IEEE Access, vol. 6, pp. 405-414.View/Download from: UTS OPUS or Publisher's site
Polysomnography (PSG) is considered the gold standard in the diagnosis of obstructive sleep apnea (OSA). The diagnosis of OSA requires an overnight sleep experiment in a laboratory. However, due to limitations in relation to the number of labs and beds available, patients often need to wait a long time before being diagnosed and eventually treated. In addition, the unfamiliar environment and restricted mobility when a patient is being tested with a polysomnogram may disturb their sleep, resulting in an incomplete or corrupted test. Therefore, it is posed that a PSG conducted in the patient's home would be more reliable and convenient. The Internet of Things (IoT) plays a vital role in the e-Health system. In this paper, we implement an IoT-based wireless polysomnography system for sleep monitoring, which utilizes a battery-powered, miniature, wireless, portable, and multipurpose recorder. A Java-based PSG recording program in the personal computer is designed to save several bio-signals and transfer them into the European data format. These PSG records can be used to determine a patient's sleep stages and diagnose OSA. This system is portable, lightweight, and has low power-consumption. To demonstrate the feasibility of the proposed PSG system, a comparison was made between the standard PSG-Alice 5 Diagnostic Sleep System and the proposed system. Several healthy volunteer patients participated in the PSG experiment and were monitored by both the standard PSG-Alice 5 Diagnostic Sleep System and the proposed system simultaneously, under the supervision of specialists at the Sleep Laboratory in Taipei Veteran General Hospital. A comparison of the results of the time-domain waveform and sleep stage of the two systems shows that the proposed system is reliable and can be applied in practice. The proposed system can facilitate the long-Term tracing and research of personal sleep monitoring at home.
Ming, Y, Pelusi, D, Fang, CN, Prasad, M, Wang, YK, Wu, D & Lin, CT 2018, 'EEG data analysis with stacked differentiable neural computers', Neural Computing and Applications.View/Download from: UTS OPUS or Publisher's site
© 2018, Springer-Verlag London Ltd., part of Springer Nature. Differentiable neural computer (DNC) has demonstrated remarkable capabilities in solving complex problems. In this paper, we propose to stack an enhanced version of differentiable neural computer together to extend its learning capabilities. Firstly, we give an intuitive interpretation of DNC to explain the architectural essence and demonstrate the stacking feasibility by contrasting it with the conventional recurrent neural network. Secondly, the architecture of stacked DNCs is proposed and modified for electroencephalogram (EEG) data analysis. We substitute the original Long Short-Term Memory network controller by a recurrent convolutional network controller and adjust the memory accessing structures for processing EEG topographic data. Thirdly, the practicability of our proposed model is verified by an open-sourced EEG dataset with the highest average accuracy achieved; then after fine-tuning the parameters, we show the minimal mean error obtained on a proprietary EEG dataset. Finally, by analyzing the behavioral characteristics of the trained stacked DNCs model, we highlight the suitableness and potential of utilizing stacked DNCs in EEG signal processing.
El-Sayed, H, Sankar, S, Daraghmi, Y-A, Tiwari, P, Rattagan, E, Mohanty, M, Puthal, D & Prasad, M 2018, 'Accurate Traffic Flow Prediction in Heterogeneous Vehicular Networks in an Intelligent Transport System Using a Supervised Non-Parametric Classifier.', Sensors (Basel, Switzerland), vol. 18, no. 6.View/Download from: UTS OPUS or Publisher's site
Heterogeneous vehicular networks (HETVNETs) evolve from vehicular ad hoc networks (VANETs), which allow vehicles to always be connected so as to obtain safety services within intelligent transportation systems (ITSs). The services and data provided by HETVNETs should be neither interrupted nor delayed. Therefore, Quality of Service (QoS) improvement of HETVNETs is one of the topics attracting the attention of researchers and the manufacturing community. Several methodologies and frameworks have been devised by researchers to address QoS-prediction service issues. In this paper, to improve QoS, we evaluate various traffic characteristics of HETVNETs and propose a new supervised learning model to capture knowledge on all possible traffic patterns. This model is a refinement of support vector machine (SVM) kernels with a radial basis function (RBF). The proposed model produces better results than SVMs, and outperforms other prediction methods used in a traffic context, as it has lower computational complexity and higher prediction accuracy.
Kaiwartya, O, Abdullah, AH, Cao, Y, Lloret, J, Kumar, S, Shah, RR, Prasad, M & Prakash, S 2018, 'Virtualization in Wireless Sensor Networks: Fault Tolerant Embedding for Internet of Things', IEEE Internet of Things Journal, vol. 5, no. 2, pp. 571-580.View/Download from: UTS OPUS or Publisher's site
IEEE Recently, virtualization in wireless sensor networks (WSNs) has witnessed significant attention due to the growing service domain for IoT. Related literature on virtualization in WSNs explored resource optimization without considering communication failure in WSNs environments. The failure of a communication link in WSNs impacts many virtual networks running IoT services. In this context, this paper proposes a framework for optimizing fault tolerance in virtualization in WSNs, focusing on heterogeneous networks for service-oriented IoT applications. An optimization problem is formulated considering fault tolerance and communication delay as two conflicting objectives. An adapted non-dominated sorting based genetic algorithm (A-NSGA) is developed to solve the optimization problem. The major components of A-NSGA include chromosome representation, fault tolerance and delay computation, crossover and mutation, and non-dominance based sorting. Analytical and simulation based comparative performance evaluation has been carried out. From the analysis of results, it is evident that the framework effectively optimizes fault tolerance for virtualization in WSNs.
Lenka, RK, Rath, AK, Tan, Z, Sharma, S, Puthal, D, Simha, NVR, Prasad, M, Raja, R & Tripathi, SS 2018, 'Building Scalable Cyber-Physical-Social Networking Infrastructure Using IoT and Low Power Sensors', IEEE Access, vol. 6, pp. 30162-30173.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. Wireless sensors are an important component to develop the Internet of Things (IoT) Sensing infrastructure. There are enormous numbers of sensors connected with each other to form a network (well known as wireless sensor networks) to complete the IoT Infrastructure. These deployed wireless sensors are with limited energy and processing capabilities. The IoT infrastructure becomes a key factor to building cyber-physical-social networking infrastructure, where all these sensing devices transmit data toward the cloud data center. Data routing toward cloud data center using such low power sensor is still a challenging task. In order to prolong the lifetime of the IoT sensing infrastructure and building scalable cyber infrastructure, there is the requirement of sensing optimization and energy efficient data routing. Toward addressing these issues of IoT sensing, this paper proposes a novel rendezvous data routing protocol for low-power sensors. The proposed method divides the sensing area into a number of clusters to lessen the energy consumption with data accumulation and aggregation. As a result, there will be less amount of data stream to the network. Another major reason to select cluster-based data routing is to reduce the control overhead. Finally, the simulation of the proposed method is done in the Castalia simulator to observe the performance. It has been concluded that the proposed method is energy efficient and it prolongs the networks lifetime for scalable IoT infrastructure.
Puthal, D, Obaidat, MS, Nanda, P, Prasad, M, Mohanty, SP & Zomaya, AY 2018, 'Secure and Sustainable Load Balancing of Edge Data Centers in Fog Computing', IEEE Communications Magazine, vol. 56, no. 5, pp. 60-65.View/Download from: UTS OPUS or Publisher's site
Fog computing is a recent research trend to bring cloud computing services to network edges. EDCs are deployed to decrease the latency and network congestion by processing data streams and user requests in near real time. EDC deployment is distributed in nature and positioned between cloud data centers and data sources. Load balancing is the process of redistributing the work load among EDCs to improve both resource utilization and job response time. Load balancing also avoids a situation where some EDCs are heavily loaded while others are in idle state or doing little data processing. In such scenarios, load balancing between the EDCs plays a vital role for user response and real-time event detection. As the EDCs are deployed in an unattended environment, secure authentication of EDCs is an important issue to address before performing load balancing. This article proposes a novel load balancing technique to authenticate the EDCs and find less loaded EDCs for task allocation. The proposed load balancing technique is more efficient than other existing approaches in finding less loaded EDCs for task allocation. The proposed approach not only improves efficiency of load balancing; it also strengthens the security by authenticating the destination EDCs.
Prasad, M, Lin, CT, Li, DL, Hong, CT, Ding, WP & Chang, JY 2017, 'Soft-Boosted Self-Constructing Neural Fuzzy Inference Network', IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, vol. 47, no. 3, pp. 584-588.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. This correspondence paper proposes an improved version of the self-constructing neural fuzzy inference network (SONFIN), called soft-boosted SONFIN (SB-SONFIN). The design softly boosts the learning process of the SONFIN in order to decrease the error rate and enhance the learning speed. The SB-SONFIN boosts the learning power of the SONFIN by taking into account the numbers of fuzzy rules and initial weights which are two important parameters of the SONFIN, SB-SONFIN advances the learning process by: 1) initializing the weights with the width of the fuzzy sets rather than just with random values and 2) improving the parameter learning rates with the number of learned fuzzy rules. The effectiveness of the proposed soft boosting scheme is validated on several real world and benchmark datasets. The experimental results show that the SB-SONFIN possesses the capability to outperform other known methods on various datasets.
Prasad, M, Liu, YT, Li, DL, Lin, CT, Shah, RR & Kaiwartya, OP 2017, 'A new mechanism for data visualization with TSK-type preprocessed collaborative fuzzy rule based system', Journal of Artificial Intelligence and Soft Computing Research, vol. 7, no. 1, pp. 33-46.View/Download from: UTS OPUS or Publisher's site
Saxena, A, Prasad, M, Gupta, A, Bharill, N, Patel, OP, Tiwari, A, Er, MJ, Ding, W & Lin, CT 2017, 'A review of clustering techniques and developments', Neurocomputing, vol. 267, pp. 664-681.View/Download from: UTS OPUS or Publisher's site
© 2017 Elsevier B.V. This paper presents a comprehensive study on clustering: exiting methods and developments made at various times. Clustering is defined as an unsupervised learning where the objects are grouped on the basis of some similarity inherent among them. There are different methods for clustering the objects such as hierarchical, partitional, grid, density based and model based. The approaches used in these methods are discussed with their respective states of art and applicability. The measures of similarity as well as the evaluation criteria, which are the central components of clustering, are also presented in the paper. The applications of clustering in some fields like image segmentation, object and character recognition and data mining are highlighted.
Abdul Hanan, AH, Yazid Idris, M, Kaiwartya, O, Prasad, M & Ratn Shah, R 2017, 'Real traffic-data based evaluation of vehicular traffic environment and state-of-the-art with future issues in location-centric data dissemination for VANETs', Digital Communications and Networks, vol. 3, no. 3, pp. 195-210.View/Download from: UTS OPUS or Publisher's site
© 2017 Chongqing University of Posts and Telecommuniocations Extensive investigation has been performed in location-centric or geocast routing protocols for reliable and efficient dissemination of information in Vehicular Adhoc Networks (VANETs). Various location-centric routing protocols have been suggested in literature for road safety ITS applications considering urban and highway traffic environment. This paper characterizes vehicular environments based on real traffic data and investigates the evolution of location-centric data dissemination. The current study is carried out with three main objectives: (i) to analyze the impact of dynamic traffic environment on the design of data dissemination techniques, (ii) to characterize location-centric data dissemination in terms of functional and qualitative behavior of protocols, properties, and strengths and weaknesses, and (iii) to find some future research directions in information dissemination based on location. Vehicular traffic environments have been classified into three categories based on physical characteristics such as speed, inter-vehicular distance, neighborhood stability, traffic volume, etc. Real traffic data is considered to analyze on-road traffic environments based on the measurement of physical parameters and weather conditions. Design issues are identified in incorporating physical parameters and weather conditions into data dissemination. Functional and qualitative characteristics of location-centric techniques are explored considering urban and highway environments. Comparative analysis of location-centric techniques is carried out for both urban and highway environments individually based on some unique and common characteristics of the environments. Finally, some future research directions are identified in the area based on the detailed investigation of traffic environments and location-centric data dissemination techniques.
Kaiwartya, O, Prasad, M, Prakash, S, Samadhiya, D, Abdullah, AH & Rahman, SOA 2017, 'An investigation on biometric internet security', International Journal of Network Security, vol. 19, no. 2, pp. 167-176.View/Download from: UTS OPUS or Publisher's site
Due to the Internet revolution in the last decade, each and every work area of society are directly or indirectly depending on computers, highly integrated computer networks and communication systems, electronic data storage and high transfer based devices, e-commerce, e-security, e-governance, and e-business. The Internet revolution is also emerged as significant challenge due to the threats of hacking systems and individual accounts, malware, fraud and vulnerabilities of system and networks, etc. In this context, this paper explores E-Security in terms of challenges and measurements. Biometric recognition is also investigated as a key e-security solution. E-Security is precisely described to understand the concept and requirements. The major challenges of e-security; namely, threats, attacks, vulnerabilities are presented in detail. Some measurement are identified and discussed for the challenges. Biometric recognition is discussed in detail wit pros and cons of the approach as a key e-security solution. This investigation helps in clear understating of e-security challenges and possible implementation of the identified measurements for the challenges in wide area of network communications.
Sharma, S, Puthal, D, Tazeen, S, Prasad, M & Zomaya, AY 2017, 'MSGR: A Mode-Switched Grid-Based Sustainable Routing Protocol for Wireless Sensor Networks', IEEE Access, vol. 5, pp. 19864-19875.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. A Wireless Sensor Network (WSN) consists of enormous amount of sensor nodes. These sensor nodes sense the changes in physical parameters from the sensing range and forward the information to the sink nodes or the base station. Since sensor nodes are driven with limited power batteries, prolonging the network lifetime is difficult and very expensive, especially for hostile locations. Therefore, routing protocols for WSN must strategically distribute the dissipation of energy, so as to increase the overall lifetime of the system. Current research trends from areas, such as from Internet of Things and fog computing use sensors as the source of data. Therefore, energy-efficient data routing in WSN is still a challenging task for real-Time applications. Hierarchical grid-based routing is an energy-efficient method for routing of data packets. This method divides the sensing area into grids and is advantageous in wireless sensor networks to enhance network lifetime. The network is partitioned into virtual equal-sized grids. The proposed mode-switched grid-based routing protocol for WSN selects one node per grid as the grid head. The routing path to the sink is established using grid heads. Grid heads are switched between active and sleep modes alternately. Therefore, not all grid heads take part in the routing process at the same time. This saves energy in grid heads and improves the network lifetime. The proposed method builds a routing path using each active grid head which leads to the sink. For handling the mobile sink movement, the routing path changes only for some grid head nodes which are nearer to the grid, in which the mobile sink is currently positioned. Data packets generated at any source node are routed directly through the data disseminating grid head nodes on the routing path to the sink.
Ding, W-P, Lin, C-T, Prasad, M, Chen, S-B & Guan, Z-J 2016, 'Attribute Equilibrium Dominance Reduction Accelerator (DCCAEDR) Based on Distributed Coevolutionary Cloud and Its Application in Medical Records', IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, vol. 46, no. 3, pp. 384-400.View/Download from: UTS OPUS or Publisher's site
Kaiwartya, O, Abdullah, AH, Cao, Y, Altameem, A, Prasad, M, Lin, CT & Liu, X 2016, 'Internet of Vehicles: Motivation, Layered Architecture, Network Model, Challenges, and Future Aspects', IEEE Access, vol. 4, pp. 5356-5373.View/Download from: UTS OPUS or Publisher's site
© 2013 IEEE. Internet of Things is smartly changing various existing research areas into new themes, including smart health, smart home, smart industry, and smart transport. Relying on the basis of 'smart transport,' Internet of Vehicles (IoV) is evolving as a new theme of research and development from vehicular ad hoc networks (VANETs). This paper presents a comprehensive framework of IoV with emphasis on layered architecture, protocol stack, network model, challenges, and future aspects. Specifically, following the background on the evolution of VANETs and motivation on IoV an overview of IoV is presented as the heterogeneous vehicular networks. The IoV includes five types of vehicular communications, namely, vehicle-to-vehicle, vehicle-to-roadside, vehicle-to-infrastructure of cellular networks, vehicle-to-personal devices, and vehicle-to-sensors. A five layered architecture of IoV is proposed considering functionalities and representations of each layer. A protocol stack for the layered architecture is structured considering management, operational, and security planes. A network model of IoV is proposed based on the three network elements, including cloud, connection, and client. The benefits of the design and development of IoV are highlighted by performing a qualitative comparison between IoV and VANETs. Finally, the challenges ahead for realizing IoV are discussed and future aspects of IoV are envisioned.
This paper, proposes a novel artificial neural network, called self-adjusting feature map (SAM), and develop its unsupervised learning ability with self-adjusting mechanism. The trained network structure of representative connected neurons not only displays the spatial relation of the input data distribution but also quantizes the data well. The SAM can automatically isolate a set of connected neurons, in which, the used number of the sets may indicate the number of clusters. The idea of self-adjusting mechanism is based on combining of mathematical statistics and neurological advantages and retreat of waste. In the training process, for each representative neuron has are three phases, growth, adaptation and decline. The network of representative neurons, first create the necessary neurons according to the local density of the input data in the growth phase. In the adaption phase, it adjusts neighborhood neuron pair's connected/disconnected topology constantly according to the statistics of input feature data. Finally, the unnecessary neurons of the network are merged or remove in the decline phase. In this paper, we exploit the SAM to handle some peculiar cases that cannot be handled easily by classical unsupervised learning networks such as self-organizing map (SOM) network. The remarkable characteristics of the SAM can be seen on various real world cases in the experimental results.
Singh, J, Prasad, M, Prasad, OK, Joo, EM, Saxena, AK & Lin, C-T 2016, 'A Novel Fuzzy Logic Model for Pseudo-Relevance Feedback-Based Query Expansion', INTERNATIONAL JOURNAL OF FUZZY SYSTEMS, vol. 18, no. 6, pp. 980-989.View/Download from: UTS OPUS or Publisher's site
Lin, C-T, Prasad, M & Saxena, A 2015, 'An Improved Polynomial Neural Network Classifier Using Real-Coded Genetic Algorithm', IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, vol. 45, no. 11, pp. 1389-1401.View/Download from: UTS OPUS or Publisher's site
Prasad, M, Li, DL, Lin, C, Prakash, S, Singh, J & Joshi, S 2015, 'Designing Mamdani-Type Fuzzy Reasoning for Visualizing Prediction Problems Based on Collaborative Fuzzy Clustering', IAENG International Journal of Computer Science, vol. 42, no. 4, pp. 404-411.View/Download from: UTS OPUS
In this paper a collaborative fuzzy c-means (CFCM) is used to generate fuzzy rules for fuzzy inference systems to evaluate the time series model. CFCM helps system to integrate two or more different datasets having similar features which are collected at the different environment with the different time period and it integrates these datasets together in order to visualize some common patterns among the datasets. In order to do any mode of integration between datasets, there is a necessity to define the common features between datasets by using some kind of collaborative process and also need to preserve the privacy and security at higher levels. This collaboration process gives a common structure between datasets which helps to define an appropriate number of rules for structural learning and also improve the accuracy of the system modeling
Prasad, M, Lin, YY, Lin, CT, Er, MJ & Prasad, OK 2015, 'A new data-driven neural fuzzy system with collaborative fuzzy clustering mechanism', NEUROCOMPUTING, vol. 167, pp. 558-568.View/Download from: UTS OPUS or Publisher's site
Za'in, C, Pratama, M, Prasad, M, Puthal, D, Lim, CP & Seera, M 2018, 'Motor fault detection and diagnosis based on a meta-cognitive random vector functional link network' in Fault Diagnosis of Hybrid Dynamic and Complex Systems, Springer, Switzerland, pp. 15-44.View/Download from: UTS OPUS or Publisher's site
Thiyagarajan, K, Kodagoda, S, Ulapane, N & Prasad, M 2020, 'A Temporal Forecasting Driven Approach Using Facebook’s Prophet Method for Anomaly Detection in Sewer Air Temperature Sensor System', 2020 15th IEEE Conference on Industrial Electronics and Applications, Kristiansand, Norway.View/Download from: UTS OPUS
Cao, Z, Chang, YC, Prasad, M, Tanveer, M & Lin, CT 2019, 'Tensor decomposition for EEG signals retrieval', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2423-2427.View/Download from: Publisher's site
© 2019 IEEE. Prior studies have proposed methods to recover multi-channel electroencephalography (EEG) signal ensembles from their partially sampled entries. These methods depend on spatial scenarios, yet few approaches aiming to a temporal reconstruction with lower loss. The goal of this study is to retrieve the temporal EEG signals independently which was overlooked in data pre-processing. We considered EEG signals are impinging on tensor-based approach, named nonlinear Canonical Polyadic Decomposition (CPD). In this study, we collected EEG signals during a resting-state task. Then, we defined that the source signals are original EEG signals and the generated tensor is perturbed by Gaussian noise with a signal-to-noise ratio of 0 dB. The sources are separated using a basic nonnegative CPD and the relative errors on the estimates of the factor matrices. Comparing the similarities between the source signals and their recovered versions, the results showed significantly high correlation over 95%. Our findings reveal the possibility of recoverable temporal signals in EEG applications.
Qararyah, F, Daraghmi, YA, Daraghmi, E, Rajora, S, Lin, CT & Prasad, M 2018, 'A Time Efficient Model for Region of Interest Extraction in Real Time Traffic Signs Recognition System', Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018, IEEE Symposium Series on Computational Intelligence, IEEE, India, pp. 83-87.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. Computation intelligence plays a major role in developing intelligent vehicles, which contains a Traffic Sign Recognition (TSR) system for increasing vehicle safety. Traffic sign recognition systems consist of an initial phase called Traffic Sign Detection (TSD), where images and colors are segmented and fed to the recognition phase. The most challenging process in TSR systems in terms of time consumption is the detection phase. The previous studies proposed different models for traffic sign detection, however, the computation time of these models still requires improvement for enabling real time systems. Therefore, this paper focuses on the computational time and proposes a novel time efficient color segmentation model based on logistic regression. This paper uses RGB color space as the domain to extract the features of our hypothesis; this has boosted the speed of the proposed model, since no color conversion is needed. The trained segmentation classifier is tested on 1000 traffic sign images taken in different lighting conditions. The experimental results show that the proposed model segmented 974 of these images correctly and in a time less than one-fifth of the time needed by any other robust segmentation methods.
Sawhney, R, Shah, RR, Bhatia, V, Lin, CT, Aggarwal, S & Prasad, M 2019, 'Exploring the Impact of Evolutionary Computing based Feature Selection in Suicidal Ideation Detection', IEEE International Conference on Fuzzy Systems.View/Download from: UTS OPUS or Publisher's site
© 2019 IEEE. The ubiquitous availability of smartphones and the increasing popularity of social media provide a platform for users to express their feelings, including suicidal ideation. Suicide prevention by suicidal ideation detection on social media lights the path to controlling the rapidly increasing suicide rates amongst youth. This paper proposes a diverse set of features and investigates into feature selection using the Firefly algorithm to build an efficient and robust supervised approach to classifying tweets with suicidal ideation. The development of a suicidal language to create three diverse, manually annotated datasets leads to the validation of the proposed model. An in-depth result and error analysis lead to an accurate system for monitoring suicidal ideation on social media along with the discovery of optimal feature subsets and selection methods using a penalty based Firefly algorithm.
Borah, P, Gupta, D & Prasad, M 2018, 'Improved 2-norm Based Fuzzy Least Squares Twin Support Vector Machine', Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018, IEEE Symposium Series on Computational Intelligence, IEEE, Bangalore, India, pp. 412-419.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. In order to reduce the higher training cost of support vector machine (SVM) and its sensitivity towards noise and outliers, two fuzzy based approaches are proposed in this paper. The proposed approaches are based on least squares twin support vector machine (LSTWSVM) and fuzzy support vector machine (FSVM). The effects of noise and outliers are reduced by assigning lower membership values to the data points which are away from the class centers. Further, 2-norm of the slack vectors of the LSTWSVM formulation is taken after multiplying to their respective diagonal matrices of the membership values to effectively utilize the fuzzy membership principle and to make the optimization problem strongly convex. Moreover, the proposed approaches solve linear equations instead of quadratic programming problems which help in training faster. The effectiveness of the proposed approaches are established by comparing the classification accuracies and training time with support vector machine, fuzzy support vector machine, twin support vector machine and least squares twin support vector machine.
Gupta, D, Sarma, HJ, Mishra, K & Prasad, M 2019, 'Regularized universum twin support vector machine for classification of EEG signal', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2298-2304.View/Download from: Publisher's site
© 2019 IEEE. Electroencephalogram signal is the signal used for the detection of a neurological disorder as epilepsy disorder, sleep disorder and many more. The types of EEG signal gives the hidden information regarding the distribution of the data that may consist of a large volume of the poor and noisy signal. In order to reduce the outlier effects and noise, incorporation of prior knowledge in the model, universum may help and enhance the better generalization ability of the model. This paper proposes a regularized universum twin support vector machine (RUTWSVM) for classification of the healthy and seizure EEG signals. Here, the selection of the universum data points is obtained in two ways (i). Universum data has been generated from the healthy and seizure EEG signals itself and (ii). Interictal EEG signal has been used as universum data which may help to handle the outlier effects. Further, various feature selection techniques are applied to extract the important noise free features from the EEG signals. We have performed a comparative analysis of proposed RUTWSVM with USVM and UTWSVM to classify the EEG signals as well as benchmark real-world datasets in an optimum way. The experiment results clearly exhibit the applicability and usability of the proposed RUTWSVM with interictal EEG signals as universum data points as well as benchmark real-world datasets.
Gupta, U, Gupta, D & Prasad, M 2018, 'Kernel Target Alignment based Fuzzy Least Square Twin Bounded Support Vector Machine', Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018, IEEE Symposium Series on Computational Intelligence, IEEE, Bangalore, India, pp. 228-235.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. A kernel-target alignment based fuzzy least square twin bounded support vector machine (KTAFLSTBSVM) is proposed to reduce the effects of outliers and noise. The proposed model is an effective and efficient fuzzy based least square twin bounded support vector machine for binary classification where the membership values are assigned based on kernel-target alignment approach. The proposed KTA-FLSTBSVM solves the two systems of linear equations, which is computationally very fast with significant comparable performance. To development the robust model, this approach minimizes the structural risk which is the gist of statistical learning theory. This powerful KTA-FLSTBSVM approach is tested on artificial data sets as well as benchmark real-world datasets that provide significantly better result in terms of generalization performance and computational time.
Jiang, S, Cao, J & Prasad, M 2019, 'The Metrics to Evaluate the Health Status of OSS Projects Based on Factor Analysis', Communications in Computer and Information Science, pp. 723-737.View/Download from: Publisher's site
© 2019, Springer Nature Singapore Pte Ltd. As open-source software (OSS) development is becoming a trend, an increasing number of businesses and developers are joining OSS projects. For project managers, developers and users, understanding the current health status of a project is very important to manage a development process, select the open-source projects to development or to adopt the software packages developed by projects. Therefore, an efficient approach to evaluate the health status of the open-source project is needed. Unfortunately, although many approaches including metrics have been proposed, they are designed in arbitrary ways. In this paper, a math ematical tool, i.e., factor analysis, is used to build a health evaluation model for OSS projects. As far as we know, this is the first time that factor analysis has been applied to evaluate OSS projects. This model is based on GitHub data and uses the basic indexes that are closely related to the health status of the projects as the input. Then, six new synthetic metrics, namely community activity, project popularity, development activity, completeness, responsiveness and persistence are obtained through factor analysis, which can be used to calculate the overall health score of a project. Moreover, in order to verify the effectiveness of this model, it is applied to some real projects and the results show that the overall scores achieved by this model can reflect the health status of the projects.
Meena, MS, Singh, P, Rana, A, Mery, D & Prasad, M 2019, 'A Robust Face Recognition System for One Sample Problem', Image and Video Technology (LNCS), Pacific-Rim Symposium on Image and Video Technology, Springer, Sydney, NSW, Australia, pp. 13-26.View/Download from: UTS OPUS or Publisher's site
© 2019, Springer Nature Switzerland AG. Most of the practical applications have limited number of image samples of individuals for face verification and recognition process such as passport, driving licenses, photo ID etc. So use of computer system becomes challenging task, when image samples available per person for training and testing of system are limited. We are proposing a robust face recognition system based on Tetrolet, Local Directional Pattern (LDP) and Cat Swam Optimization (CSO) to solve this problem. Initially, the input image is pre-processed to extract region of interest using filtering method. This image is then given to the proposed descriptor, namely Tetrolet-LDP to extract the features of the image. The features are subjected to classification using the proposed classification module, called Cat Swarm Optimization based 2-Dimensional Hidden Markov Model (CSO-based 2DHMM) in which the CSO trains the 2D-HMM. The performance is analyzed using the metrics, such as accuracy, False Rejection Rate (FRR), & False Acceptance Rate (FAR) and the system achieves high accuracy of 99.65%, and less FRR and FAR of 0.0033 and 0.003 for training percentage variation and 99.65%, 0.0035 and 0.004 for k-Fold Validation.
Prasad, M, Rajora, S, Gupta, D, Daraghmi, YA, Daraghmi, E, Yadav, P, Tiwari, P & Saxena, A 2019, 'Fusion based En-FEC Transfer Learning Approach for Automobile Parts Recognition System', Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018, IEEE Symposium Series on Computational Intelligence, IEEE, Bangalore, India,, pp. 2193-2199.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. The artificially supervised classification of real world entities have gained a phenomenal significance in recent year of computational advancements. An intelligent classification model focuses on rendering accurate outcomes vide the implicated paradigms with respect to the subjected data employed to train the classifier. This paper proposes a novel deep learning approach to classify the various parts of any operational engine such as crank shafts, rock-arms, distributer, air duct, assecorybelt etc. Deployed in automobiles. The proposed architecture distinctively utilizes convolution neural networks for this typical classification problem and altogether constructs a robust transfer learning paradigm to render the correct class label against the validation and test images as the conclusive result of the classification. The proposed methodology poses in such a way that it can qualitatively classify and henceforth give the corresponding class label of the machinery/engine part under consideration. This computationally intelligent architecture requires the user to feed the image of the engine part to the model in order to achieve the requisite responses of classification. The main contribution of the proposed method is the development of a robust algorithm that can exhibit pronounced results without training the entire ConvNet architecture from scratch, thereby enabling the proposed paradigm to be deployable in application instances wherein limited labeled training data is available.
Rahman, MA, Singh, P, Muniyandi, RC, Mery, D & Prasad, M 2019, 'Prostate Cancer Classification Based on Best First Search and Taguchi Feature Selection Method', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 325-336.View/Download from: Publisher's site
© 2019, Springer Nature Switzerland AG. Prostate cancer is the second most common cancer occurring in men worldwide, about 1 in 41 men will die because of prostate cancer. Death rates of prostate cancer increases with age. Even though, it being a serious condition only about 1 man in 9 will be diagnosed with prostate cancer during his lifetime. Accurate and early diagnosis can help clinician to treat the cancer better and save lives. This paper proposes two phases feature selection method to enhance prostate cancer early diagnosis based on artificial neural network. In the first phase, Best First Search method is used to extract the relevant features from original dataset. In the second phase, Taguchi method is used to select the most important feature from the already extracted features from Best First Search method. A public available prostate cancer benchmark dataset is used for experiment, which contains two classes of data normal and abnormal. The proposed method outperforms other existing methods on prostate cancer benchmark dataset with classification accuracy of 98.6%. The proposed approach can help clinicians to reach at more accurate and early diagnosis of different stages of prostate cancer and so that they make most suitable treatment decision to save lives of patients and prevent death due to prostate cancer.
Rajora, S, Kumar Vishwakarma, D, Singh, K & Prasad, M 2018, 'CSgI: A Deep Learning based approach for Marijuana Leaves Strain Classification', 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2018, Annual Information Technology, Electronics and Mobile Communication Conference, IEEE, Vancouver, BC, Canada, pp. 209-214.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. This paper proposes a novel approach that classifies the images of various marijuana/cannabis leaves into their respective classes of strains and types. The proposed architecture works on a two-fold technique which when implemented in the requisite sequence delivers phenomenal results to the classification problem statement. The first fold, being the segmentation or foreground extraction in the images, focuses on extracting the RDI (Region of Interest) using a robust segmentation algorithm which can suitable separate the foreground from the image; and the second fold, being the Deep Learning aspect focuses on the result classification task. This literature gives a quantitative analysis of implementing this classification problem vide a transfer learning paradigm (for application instances with less training data in hand) training the entire CNN archetype from scratch (for application instances with sufficient training data in hand). Thus, altogether the proposed methodology distinctively deploys ConvNets for the posed classification problem having dual aspects of approaches implementation wiz: a) Transfer Learning b) Training the entire CNN from scratch. The novelty of the proposed work can be counted upon as the construction of a robust algorithm very first of its kind in this respective application domain which is potent enough to render the correct class label of the strain/type of marijuana or cannabis leaf image when fed to the system for classification.
Rajora, S, Li, DL, Jha, C, Bharill, N, Patel, OP, Joshi, S, Puthal, D & Prasad, M 2018, 'A Comparative Study of Machine Learning Techniques for Credit Card Fraud Detection Based on Time Variance', Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018, IEEE Symposium Series on Computational Intelligence, IEEE, Bangalore, India, India, pp. 1958-1963.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. This paper proposes a comparative performance of ten different machine learning algorithms, done on a credit card fraud detection application. The machine learning methods have been classified into two groups namely classification algorithms and ensemble learning group. Each group is comprised of five different algorithms. Besides, the 'Time' feature is introduced in the data set and performances of the algorithms are studied with and without the 'Time' feature. Two algorithms of the ensemble learning group have been found to perform better when the used dataset does not include the 'Time' feature. However, for the classification algorithms group, three classifiers are found to show better predictive accuracies when all attributes are included in the used dataset. The rest of the machine learning models have approximate similar scores between these datasets.
Soomro, AM, Paryani, S, Rehman, J, Echeverria, R, Prasad, M & Biloria, N 2019, 'Influencing Human Behaviour to Optimise Energy in Commercial Buildings', https://www.acis2019.org/papers-by-session/, Australian Conference on Information Systems 2019, ACIS 2019, Perth, Australia, pp. 901-907.View/Download from: UTS OPUS
This paper discusses the impact of user energy choices on building energy demand, and how energy choices could be influenced to minimise building energy consumption using information systems.
Accordingly, a socio-technical framework is designed and presented, which draws upon the use of energy interventions. A novel Social-Economic-Environmental (SEE) model is presented within the sociotechnical framework which is aimed at nudging inhabitants enabling them to conserve energy in the university buildings, thereby making the world a sustainable place to live. The framework takes into account the Agent-based Modelling (ABM) approach to model user energy choices and their willingness to conserve energy in buildings. This research intends to test the socio-technical framework in the next stage of this study. Finally, this paper highlights gaps and the significance of understanding how user behaviour and their energy consumption can be influenced to optimise energy in university buildings, thereby reducing global greenhouse emissions.
Tiwary, M, Sharma, S, Mishra, P, El-Sayed, H, Prasad, M & Puthal, D 2018, 'Building Scalable Mobile Edge Computing by Enhancing Quality of Services', 2018 International Conference on Innovations in Information Technology (IIT), International Conference on Innovations in Information Technology, IEEE, Al Ain, United Arab Emirates, pp. 141-146.View/Download from: UTS OPUS or Publisher's site
Fu, L, Li, J, Zhou, L, Ma, Z, Liu, S, Lin, Z & Prasad, M 2018, 'Utilizing Information from Task-Independent Aspects via GAN-Assisted Knowledge Transfer', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Rio de Janeiro, Brazil.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. Observed data often have multiple labels with respect to different aspects. For example, a picture can have one label specifying the contents in terms of the object category such as aeroplane, building, cat, etc. And in the meanwhile have another label describing the image style such as photo-realistic or artistic. The central idea of this work is that any annotation of the data contains precious knowledge and is not to be foregone: An analytic task focusing on one aspect of the data can benefit from the knowledge transferred from the other aspects. We propose a passive knowledge transfer scheme for deep neural network training based on the generative adversarial nets (GANs). The adversarial training scheme encourages the nets to encode data into representations that are both discriminative for the target aspect and invariant with respect to the irrelevant aspects. We show that the scheme mixes the conditional distributions of the encoded data on the irrelevant aspects, by the theory on the link between the GAN framework and the Wasserstein metric in distribution spaces. Moreover, we empirically verified the method by i) classifying images despite influence by geometric transform and ii) recognizing the movements (geometric transform) regardless the image contents.
Zhang, L, Li, J, Huang, T, Ma, Z, Lin, Z & Prasad, M 2018, 'GAN2C: Information Completion GAN with Dual Consistency Constraints', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Rio de Janeiro, Brazil.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. This paper proposes an information completion technique, GAN2C, by imposing dual consistency constraints (2C) to a closed loop encoder-decoder architecture based on the generative adversarial nets (GAN). When adopting deep neural networks as function approximators, GAN2C enables highly effective multi-modality image conversion with sparse observation in the target modes. For empirical demonstration and model evaluation, we show that trained deep neural networks in GAN2C can infer colors for grayscale images, as well as estimate rich 3D information of a scene by densely predicting the depths. The results of the experiments show that in both tasks GAN2C as a generic framework has been comparable to or advanced the state-of-the-art performance which are achieved by highly specialized systems. Code is available at https://github.com/AdalinZhang/GAN2C.
Ming, Y, Wang, YK, Prasad, M, Wu, D & Lin, CT 2018, 'Sustained Attention Driving Task Analysis based on Recurrent Residual Neural Network using EEG Data', 2018 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), IEEE International Conference on Fuzzy Systems, IEEE, Rio de Janeiro, Brazil, pp. 1-6.View/Download from: UTS OPUS or Publisher's site
This paper proposes applying recurrent residual network (RRN) for analyzing electroencephalogram (EEG) data captured during a simulated sustained attention driving task. We first address the suitableness of utilizing residual structure as well as adopting recurrent structure for EEG signal processing. Then based on these descriptions a recurrent residual network is tailored and depicted in detail. Thirdly we use an EEG dataset obtained from a sustained-attention experiment for our model justification. By applying the RRN model to the experimental data and via the competitive result achieved, we demonstrate the elegance of the proposed model. At last, we discuss the characteristics of the learned filters and their interpretations from EEG frequency band perspectives.
Prasad, M, Chang, LC, Gupta, D, Pratama, M, Sundaram, S & Lin, CT 2018, 'Online video streaming for human tracking based on weighted resampling particle filter', Procedia Computer Science, INNS Conference on Big Data and Deep Learning 2018, Bali, Indonesia, pp. 2-12.View/Download from: UTS OPUS or Publisher's site
© 2018 The Authors. Published by Elsevier Ltd. This paper proposes a weighted resampling method for particle filter which is applied for human tracking on active camera. The proposed system consists of three major parts which are human detection, human tracking, and camera control. The codebook matching algorithm is used for extracting human region in human detection system, and the particle filter algorithm estimates the position of the human in every input image. The proposed system in this paper selects the particles with highly weighted value in resampling, because it provides higher accurate tracking features. Moreover, a proportional-integral-derivative controller (PID controller) controls the active camera by minimizing difference between center of image and the position of object obtained from particle filter. The proposed system also converts the position difference into pan-tilt speed to drive the active camera and keep the human in the field of view (FOV) camera. The intensity of image changes overtime while tracking human therefore the proposed system uses the Gaussian mixture model (GMM) to update the human feature model. As regards, the temporal occlusion problem is solved by feature similarity and the resampling particles. Also, the particle filter estimates the position of human in every input frames, thus the active camera drives smoothly. The robustness of the accurate tracking of the proposed system can be seen in the experimental results.
Prasad, M, Liu, CL, Li, DL, Jha, C & Lin, CT 2018, 'Multi-view Vehicle Detection based on Part Model with Active Learning', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Rio de Janeiro, Brazil.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. Nowadays, most ofthe vehicle detection methods aim to detect only single-view vehicles, and the performance is easily affected by partial occlusion. Therefore, a novel multi-view vehicle detection system is proposed to solve the problem of partial occlusion. The proposed system is divided into two steps: Background filtering and part model. Background filtering step is used to filter out trees, sky and other road background objects. In the part model step, each of the part models is trained by samples collected by using the proposed active learning algorithm. This paper validates the performance of the background filtering method and the part model algorithm in multi-view car detection. The performance of the proposed method outperforms previously proposed methods.
Prasad, M, Zheng, DR, Mery, D, Puthal, D, Sundaram, S & Lin, CT 2018, 'A fast and self-adaptive on-line learning detection system', Procedia Computer Science, INNS Conference on Big Data and Deep Learning, Elsevier, Bali, Indonesia, pp. 13-22.View/Download from: UTS OPUS or Publisher's site
© 2018 The Authors. Published by Elsevier Ltd. This paper proposes a method to allow users to select target species for detection, generate an initial detection model by selecting a small piece of image sample and as the movie plays, continue training this detection model automatically. This method has noticeable detection results for several types of objects. The framework of this study is divided into two parts: the initial detection model and the online learning section. The detection model initialization phase use a sample size based on the proportion of users of the Haar-like features to generate a pool of features, which is used to train and select effective classifiers. Then, as the movie plays, the detecting model detects the new sample using the NN Classifier with positive and negative samples and the similarity model calculates new samples based on the fusion background model to calculate a new sample and detect the relative similarity to the target. From this relative similarity-based conservative classification of new samples, the conserved positive and negative samples classified by the video player are used for automatic online learning and training to continuously update the classifier. In this paper, the results of the test for different types of objects show the ability to detect the target by choosing a small number of samples and performing automatic online learning, effectively reducing the manpower needed to collect a large number of image samples and a large amount of time for training. The Experimental results also reveal good detection capability.
Badarinath, D, Chaitra, S, Bharill, N, Tanveer, M, Prasad, M, Suma, HN, Abhishek Appaji, M & Vinekar, A 2018, 'Study of Clinical Staging and Classification of Retinal Images for Retinopathy of Prematurity (ROP) Screening', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Rio de Janeiro, Brazil, pp. 1-6.View/Download from: UTS OPUS or Publisher's site
© 2018 IEEE. Retinopathy of Prematurity (ROP) is a disease which requires immediate precautionary measures to prevent blindness in the infants, and this condition is prevalent in premature babies in all the underdeveloped, developing, and in the developed countries as well. This paper proposes a tool by which the stage and zones of Retinopathy of Prematurity in infants can be diagnosed easily. This tool takes the input from the Retcam and detects the stage, zone, and gives a rating of 1 to 9 for classifying the severity of the disease in the infants. This is achieved by extracting the optic disc, marking the ridge, and the distance of the optic nerve. This tool can be easily used by nurses and paramedics, unlike the existing technologies which require the guidance of a specialist to come to a conclusion.
Mishra, AK, Tripathy, AK, Obaidat, MS, Tan, Z, Prasad, M, Sadoun, B & Puthal, D 2018, 'A chain topology for efficient monitoring of food grain storage using smart sensors', ICETE 2018 - Proceedings of the 15th International Joint Conference on e-Business and Telecommunications, pp. 89-98.
Copyright © 2018 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved Due to lack of an efficient monitoring system to periodically record environmental parameters for food grain storage, a huge loss of food grains in storage is reported every year in many developing countries, especially south-Asian countries. Although Smart Sensor Networks have been successfully implemented in various applications such as health-care, military, and wildlife monitoring, there are still various issues to be addressed in food grain storage monitoring applications. Due to the food grain storage infrastructure constraints, the commonly practiced network topologies of sensor devices such as mesh, star, and grid cannot provide an effective monitoring environment. In this paper, we proposed a topology using smart sensors that can effectively cover and monitor the food grain storage area. It uses a chained structure of sensor devices with directional antennas to accurately sense and report the environmental data. The proposed topology works better than common topologies due to its chain-based structure which remains unaffected by various hindrance imposed due to food grain storage infrastructure. From the experimental results it is conclude that the proposed topology has effective coverage percentage, detection accuracy, and message delivery over Cluster-based and Mesh topologies in food grain storage environments.
Prasad, M, Anh, N, Srikanth, N & Sundaram, S 2018, 'Wave Forecasting using Meta-cognitive Interval Type-2 Fuzzy Inference System', Procedia Computer Science, International Neural Network Society Conference on Big Data and Deep Learning, Elsevier, Bali, Indonesia, pp. 33-41.View/Download from: UTS OPUS or Publisher's site
Renewable energy is fast becoming a mainstay in today’s energy scenario. One of the important sources of renewable energy is the wave energy, in addition to wind, solar, tidal, etc. Wave prediction/forecasting is consequently essential in coastal and ocean engineering studies. However, it is difficult to predict wave parameters in long term and even in the short term due to its intermittent nature. This study aims to propose a solution to handle the issue using Interval type-2 fuzzy inference system, or IT2FIS. IT2FIS has been shown to be capable of handling uncertainty associated with the data. The proposed IT2FIS is a fuzzy neural network realizing Takagi-Sugeno-Kang inference mechanism employing meta-cognitive learning algorithm. The algorithm monitors knowledge in a sample to decide an appropriate learning strategy. Performance of the system is evaluated by studying significant wave heights obtained from buoys located in Singapore. The results compared with existing state-of-the art fuzzy inference system approaches clearly indicate the advantage of IT2FIS based wave prediction.
Prasad, M, Anh, N, Srikanth, N & Sundaram, S 2018, 'Wind Speed Intervals Prediction using Meta-cognitive Approach', Procedia Computer Science, International Neural Network Society Conference on Big Data and Deep Learning, Elsevier, Bali, Indonesia, pp. 23-32.View/Download from: UTS OPUS or Publisher's site
In this paper, an interval type-2 neural fuzzy inference system and its meta-cognitive learning algorithm for wind speed prediction is proposed. Interval type-2 neuro-fuzzy system is capable of handling uncertainty associated with the data and meta-cognition employs self-regulation mechanism for learning. The proposed system realizes Takagi-Sugeno-Kang inference mechanism and adopts a fast data-driven interval-reduction method. Meta-cognitive learning enables the network structure to evolve automatically based on the knowledge in data. The parameters are updated based on an extended Kalman filter algorithm. In addition, the proposed network is able to construct prediction intervals to quantify uncertainty associated with forecasts. For performance evaluation, a real-world wind speed prediction problem is utilized. Using historical data, the model provides short-term prediction intervals of wind speed. The performance of proposed algorithm is compared with existing state-of-the art fuzzy inference system approaches and the results clearly indicate its advantages in forecasting problems.
Prasad, M, Za’in, C, Pratama, M, Lughofer, E, Ferdaus, M & Cai, Q 2018, 'Big Data Analytics based on PANFIS MapReduce', Procedia Computer Science, International Neural Network Society Conference on Big Data and Deep Learning, Elsevier, Bali, Indonesia, pp. 140-152.View/Download from: UTS OPUS or Publisher's site
In this paper, a big data analytic framework is introduced for processing high-frequency data stream. This framework architecture is developed by combining an advanced evolving learning algorithm namely Parsimonious Network Fuzzy Inference System (PANFIS) with MapReduce parallel computation, where PANFIS has the capability of processing data stream in large volume. Big datasets are learnt chunk by chunk by processors in MapReduce environment and the results are fused by rule merging method, that reduces the complexity of the rules. The performance measurement has been conducted, and the results are showing that the MapReduce framework along with PANFIS evolving system helps to reduce the processing time around 22 percent in average in comparison with the PANFIS algorithm without reducing performance in accuracy.
Chou, KP, Li, DL, Prasad, M, Pratama, M, Su, SY, Lu, H, Lin, CT & Lin, WC 2017, 'Robust Facial Alignment for Face Recognition', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2017 International Conference on Neural Information Processing, Guangzhou, China, pp. 497-504.View/Download from: UTS OPUS or Publisher's site
© 2017, Springer International Publishing AG. This paper proposes a robust real-time face recognition system that utilizes regression tree based method to locate the facial feature points. The proposed system finds the face region which is suitable to perform the recognition task by geometrically analyses of the facial expression of the target face image. In real-world facial recognition systems, the face is often cropped based on the face detection techniques. The misalignment is inevitably occurred due to facial pose, noise, occlusion, and so on. However misalignment affects the recognition rate due to sensitive nature of the face classifier. The performance of the proposed approach is evaluated with four benchmark databases. The experiment results show the robustness of the proposed approach with significant improvement in the facial recognition system on the various size and resolution of given face images.
Padmanabha A G, A, Appaji M, A, Prasad, M, Lu, H & Joshi, S 2017, 'Classification of diabetic retinopathy using textural features in retinal color fundus image', Intelligent Systems and Knowledge Engineering (ISKE), 2017 12th International Conference on, International Conference on Intelligent Systems and Knowledge Engineering, IEEE, Nanjing, China, pp. 1-5.View/Download from: UTS OPUS or Publisher's site
Early, diagnosis is essential for diabetic patients
to avoid partial or complete blindness. This work presents
a new analysis method of texture features for
classification of Diabetic Retinopathy (DR). The proposed
method masks the blood vessels and optic disk segmented
and directly extracts the textural features from the
remaining retinal region. The proposed method is much
simpler with comparison of the other methods that detect
the defective regions first and then extract the required
features for classification. The Haralick texture measures
calculated are used for classification of DR. The proposed
method is evaluated through a classification of DR using
both Support Vector Machine (SVM) and Artificial
Neural Network (ANN). The results of SVM have a better
accuracy (87.5%) over ANN (79%).The performance of
the proposed method is presented also in terms of
sensitivity and specificity.
Cao, Z, Prasad, M & Lin, CT 2017, 'Estimation of SSVEP-based EEG complexity using inherent fuzzy entropy', IEEE International Conference on Fuzzy Systems, IEEE International Conference on Fuzzy Systems, IEEE, Naples, Italy.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. This study considers the dynamic changes of complexity feature by fuzzy entropy measurement and repetitive steady-state visual evoked potential (SSVEP) stimulus. Since brain complexity reflects the ability of the brain to adapt to changing situations, we suppose such adaptation is closely related to the habituation, a form of learning in which an organism decreases or increases to respond to a stimulus after repeated presentations. By a wearable electroencephalograph (EEG) with Fpz and Oz electrodes, EEG signals were collected from 20 healthy participants in one resting and five-times 15 Hz SSVEP sessions. Moreover, EEG complexity feature was extracted by multi-scale Inherent Fuzzy Entropy (IFE) algorithm, and relative complexity (RC) was defined the difference between resting and SSVEP. Our results showed the enhanced frontal and occipital RC was accompanied with increased stimulus times. Compared with the 1st SSVEP session, the RC was significantly higher than the 5th SSVEP session at frontal and occipital areas (p < 0.05). It suggested that brain has adapted to changes in stimulus influence, and possibly connected with the habituation. In conclusion, effective evaluation of IFE has a potential EEG signature of complexity in the SSEVP-based experiment.
Cheng, EJ, Prasad, M, Puthal, D, Sharma, N, Prasad, OK, Chin, PH, Lin, CT & Blumenstein, M 2017, 'Deep Learning Based Face Recognition with Sparse Representation Classification', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 665-674.View/Download from: Publisher's site
© 2017, Springer International Publishing AG. Feature extraction is an essential step in solving real-world pattern recognition and classification problems. The accuracy of face recognition highly depends on the extracted features to represent a face. The traditional algorithms uses geometric techniques, comprising feature values including distance and angle between geometric points (eyes corners, mouth extremities, and nostrils). These features are sensitive to the elements such as illumination, variation of poses, various expressions, to mention a few. Recently, deep learning techniques have been very effective for feature extraction, and deep features have considerable tolerance for various conditions and unconstrained environment. This paper proposes a two layer deep convolutional neural network (CNN) for face feature extraction and applied sparse representation for face identification. The sparsity and selectivity of deep features can strengthen sparseness for the solution of sparse representation, which generally improves the recognition rate. The proposed method outperforms other feature extraction and classification methods in terms of recognition accuracy.
Chou, KP, Li, DL, Prasad, M, Lin, CT & Lin, WC 2017, 'A method to enhance the deep learning in an aerial image', 2017 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Proceedings, International Symposium on Intelligent Signal Processing and Communication Systems, IEEE, Xiamen, China, pp. 724-728.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. In this paper, we propose a kind of pre-processing method which can be applied to the depth learning method for the characteristics of aerial image. This method combines the color and spatial information to do the quick background filtering. In addition to increase execution speed, but also to reduce the rate of false positives.
Chou, KP, Prasad, M, Gupta, D, Sankar, S, Xu, TW, Sundaram, S, Lin, CT & Lin, WC 2017, 'Block-based feature extraction model for early fire detection', 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings, IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Honolulu, HI, USA, pp. 1-6.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. Every year the fire disaster always causes a lot of casualties and property damage. Many researchers are involved in the study of related disaster prevention. Early warning systems and stable fire can significantly reduce the damage caused by fire. Many existing image-based early warning systems can perform well in a particular field. In this paper, we propose a general framework that can be applied in most realistic environments. The proposed system is based on a block-based feature extraction method, which analyses local information in separate regions leading to a reduction in computing data. Local features of fire block are extracted from the detailed characteristics of fire objects, which include fire color, fire source immobility, and disorder. Each local feature has high detection rate and filter out different false-positive cases. Global analysis with fire texture and non-moving properties are applied to further reduce false alarm rate. The proposed system is composed of algorithms with low computation. Through a series of experiments, it can be observed that Experimental results show that the proposed system has higher detection rate and low false alarm rate under various environment.
Chou, KP, Prasad, M, Li, DL, Bharill, N, Lin, YF, Hussain, F, Lin, CT & Lin, WC 2017, 'Automatic Multi-view Action Recognition with Robust Features', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Neural Information Processing, Springer, Guangzhou, China, pp. 554-563.View/Download from: UTS OPUS or Publisher's site
© 2017, Springer International Publishing AG. This paper proposes view-invariant features to address multi-view action recognition for different actions performed in different views. The view-invariant features are obtained from clouds of varying temporal scale by extracting holistic features, which are modeled to explicitly take advantage of the global, spatial and temporal distribution of interest points. The proposed view-invariant features are highly discriminative and robust for recognizing actions as the view changes. This paper proposes a mechanism for real world application which can follow the actions of a person in a video based on image sequences and can separate these actions according to given training data. Using the proposed mechanism, the beginning and ending of an action sequence can be labeled automatically without the need for manual setting. It is not necessary in the proposed approach to re-train the system if there are changes in scenario, which means the trained database can be applied in a wide variety of environments. The experiment results show that the proposed approach outperforms existing methods on KTH and WEIZMANN datasets.
Chou, KP, Prasad, M, Puthal, D, Chen, PH, Vishwakarma, DK, Sundarami, S, Lin, CT & Lin, WC 2017, 'Fast Deformable Model for Pedestrian Detection with Haar-like features', 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings, IEEE Symposium Series on Computational Intelligence, IEEE, Honolulu, HI, USA, pp. 1-8.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. This paper proposes a novel Fast Deformable Model for Pedestrian Detection (FDMPD) to detect the pedestrians efficiently and accurately in the crowded environment. Despite of multiple detection methods available, detection becomes difficult due to variety of human postures and perspectives. The proposed study is divided into two parts. First part trains six Adaboost classifiers with Haar-like feature for different body parts (e.g., head, shoulders, and knees) to build the response feature maps. Second part uses these six response feature maps with full-body model to produce spatial deep features. The combined deep features are used as an input to SVM to judge the existence of pedestrian. As per the experiments conducted on the INRIA person dataset, the proposed FDMPD approach shows greater than 44.75 % improvement compared to other state-of-the-art methods in terms of efficiency and robustness.
Hung, YC, Wang, YK, Prasad, M & Lin, CT 2017, 'Brain dynamic states analysis based on 3D convolutional neural network', Proceedings of the 2017 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2017, IEEE International Conference on Systems, Man, and Cybernetics, IEEE, Banff, AB, Canada, pp. 222-227.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. Drowsiness driving is one major factor of traffic accident. Monitoring the changes of brain signals provides an effective and direct way for drowsiness detection. One 3D convolutional neural network (3D CNN)-based forecasting system has been proposed to monitor electroencephalography (EEG) signals and predict fatigue level during driving. The limited weight sharing and channel-wise convolution were both applied to extract the significant phenomenon in various frequency bands of brain signals and the spatial information of EEG channel location, respectively. The proposed 3D CNN with limited weight sharing and channel-wise convolution has been demonstrated to predict reaction time (RT) of driving with low root mean square error (RMSE) through the brain dynamics. This proposed approach outperforms with the state-of-the-art algorithms, such as traditional CNN, Neural Network (NN), and support vector regression (SVR). Compared with traditional CNN and Artificial Neural Network, the RMSE of 3D CNN-based RT prediction has been improved 9.5% (RMSE from 0.6322 to 0.5720) and 8% (RMSE from 0.6217 to 0.5720), respectively. We envision that this study might open a new branch between deep learning application in neuro-cognitive analysis and real world application.
Gupta, D, Borah, P & Prasad, M 2017, 'A fuzzy based Lagrangian twin parametric-margin support vector machine (FLTPMSVM)', 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings, IEEE Symposium Series on Computational Intelligence, IEEE, Honolulu, HI, USA, pp. 1-7.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. In the spirit of twin parametric-margin support vector machine (TPMSVM) and support vector machine based on fuzzy membership values (FSVM), a new method termed as fuzzy based Lagrangian twin parametric-margin support vector machine (FLTPMSVM) is proposed in this paper to reduce the effect of the outliers. In FLTPMSVM, we assign the weights to each data samples on the basis of fuzzy membership values to reduce the effect of outliers. Also, we consider the square of the 2-norm of slack variables to make the objective function strongly convex and find the solution of the proposed FLTPMSVM by solving simple linearly convergent iterative schemes instead of solving a pair of quadratic programming problems as in case of SVM, TWSVM, FTSVM and TPMSVM. No need of external toolbox is required for FLTPMSVM. The numerical experiments are performed on artificial as well as well known real-world datasets which show that our proposed FLTPMSVM is having better generalization performance and less training cost in comparison to support vector machine, twin support vector machine, fuzzy twin support vector machine and twin parametric-margin support vector machine.
Singh, J, Prasad, M, Daraghmi, YA, Tiwari, P, Yadav, P, Bharill, N, Pratama, M & Saxena, A 2017, 'Fuzzy LogicHybrid model with semantic filtering approach for pseudo relevance feedback-based query expansion', 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings, Symposium Series on Computational Intelligence, IEEE, Honolulu, HI, USA, pp. 1-7.View/Download from: UTS OPUS or Publisher's site
© 2017 IEEE. Individual query expansion term selection methods have been widely investigated in an attempt to improve their performance. Each expansion term selection method has its own weaknesses and strengths. To overcome the weaknesses and utilize the strengths of individual methods, this paper combined multiple term selection methods. In this paper, initially the possibility of improving the overall performance using individual query expansion (QE) term selection methods are explored. Secondly, some well-known rank aggregation approaches are used for combining multiple QE term selection methods. Thirdly, a new fuzzy logic-based QE approach that considers the relevance score produced by different rank aggregation approaches is proposed. The proposed fuzzy logic approach combines different weights of each term using fuzzy rules to infer the weights of the additional query terms. Finally, Word2vec approach is used to filter semantically irrelevant terms obtained after applying the fuzzy logic approach. The experimental results demonstrate that the proposed approaches achieve significant improvements over each individual term selection method, aggregated method and related state-of-the-art method.
Chou, KP, Prasad, M, Lin, YY, Joshi, S, Lin, CT & Chang, JY 2014, 'Takagi-Sugeno-Kang type collaborative fuzzy rule based system', IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings, IEEE Symposium on Computational Intelligence and Data Mining, IEEE, USA, pp. 315-320.View/Download from: Publisher's site
© 2014 IEEE. In this paper, a Takagi-Sugeno-Kang (TSK) type collaborative fuzzy rule based system is proposed with the help of knowledge learning ability of collaborative fuzzy clustering (CFC). The proposed method split a huge dataset into several small datasets and applying collaborative mechanism to interact each other and this process could be helpful to solve the big data issue. The proposed method applies the collective knowledge of CFC as input variables and the consequent part is a linear combination of the input variables. Through the intensive experimental tests on prediction problem, the performance of the proposed method is as higher as other methods. The proposed method only uses one half information of given dataset for training process and provide an accurate modeling platform while other methods use whole information of given dataset for training.
Prasad, M, Chou, KP, Saxena, A, Kawrtiya, OP, Li, DL & Lin, CT 2014, 'Collaborative fuzzy rule learning for Mamdani type fuzzy inference system with mapping of cluster centers', IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CICA 2014: 2014 IEEE Symposium on Computational Intelligence in Control and Automation, Proceedings, IEEE Symposium on Computational Intelligence in Control and Automation, IEEE, USA.View/Download from: Publisher's site
© 2014 IEEE. This paper demonstrates a novel model for Mamdani type fuzzy inference system by using the knowledge learning ability of collaborative fuzzy clustering and rule learning capability of FCM. The collaboration process finds consistency between different datasets, these datasets can be generated at various places or same place with diverse environment containing common features space and bring together to find common features within them. For any kind of collaboration or integration of datasets, there is a need of keeping privacy and security at some level. By using collaboration process, it helps fuzzy inference system to define the accurate numbers of rules for structure learning and keeps the performance of system at satisfactory level while preserving the privacy and security of given datasets.
Prasad, M, Er, MJ, Lin, C-T, Prasad, OK, Mohanty, M & Singh, J 2015, 'Novel Data Knowledge Representation with TSK-type Preprocessed Collaborative Fuzzy Rule based System', 2015 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI), IEEE Symposium Series Computational Intelligence, IEEE, Cape Town, SOUTH AFRICA, pp. 14-21.View/Download from: Publisher's site
Liu, Y-T, Lin, Y-Y, Wu, S-L, Chuang, C-H, Prasad, M & Lin, C-T 2014, 'EEG-based Driving Fatigue Prediction System Using Functional-link-based Fuzzy Neural Network', PROCEEDINGS OF THE 2014 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), International Joint Conference on Neural Networks (IJCNN), IEEE, Beijing, PEOPLES R CHINA, pp. 4109-4113.
Prasad, M, Siana, L, Li, DL, Lin, CT, Liu, YT & Saxena, A 2014, 'A preprocessed induced partition matrix based collaborative fuzzy clustering for data analysis', 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), IEEE International Conference on Fuzzy Systems, IEEE, Beijing, China, pp. 1553-1558.View/Download from: UTS OPUS or Publisher's site
© 2014 IEEE. Preprocessing is generally used for data analysis in the real world datasets that are noisy, incomplete and inconsistent. In this paper, preprocessing is used to refine the inconsistency of the prototype and partition matrices before getting involved in the collaboration process. To date, almost all organizations are trying to establish some collaboration with others in order to enhance the performance of their services. Due to privacy and security issues they cannot share their information and data with each other. Collaborative clustering helps this kind of collaborative process while maintaining the privacy and security of data and can still yield a satisfactory result. Preprocessing helps the collaborative process by using an induced partition matrix generated based on cluster prototypes. The induced partition matrix is calculated from local data by using the cluster prototypes obtained from other data sites. Each member of the collaborating team collects the data and generates information locally by using the fuzzy c-means (FCM) and shares the cluster prototypes to other members. The other members preprocess the centroids before collaboration and use this information to share globally through collaborative fuzzy clustering (CFC) with other data. This process helps system to learn and gather information from other data sets. It is found that preprocessing helps system to provide reliable and satisfactory result, which can be easily visualized through our simulation results in this paper.
Lin, C-T, Prasad, M & Chang, J-Y 2013, 'Designing Mamdani Type Fuzzy Rule Using a Collaborative FCM Scheme', 2013 INTERNATIONAL CONFERENCE ON FUZZY THEORY AND ITS APPLICATIONS (IFUZZY 2013), International Conference on Fuzzy Theory and Its Applications (iFUZZY), IEEE, Taipei, TAIWAN, pp. 279-282.
Prasad, M, Lin, CT, Yang, CT & Saxena, A 2013, 'Vertical collaborative fuzzy C-means for multiple EEG data sets', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Intelligent Robotics and Applications, Springer, Busan, South Korea, pp. 246-257.View/Download from: UTS OPUS or Publisher's site
Vertical Collaborative Fuzzy C-Means (VC-FCM) is a clustering method that performs clustering on a data set of having some set of patterns with the collaboration of some knowledge which is obtained from other data set having the same number of features but different set of patterns. Uncertain relationship lies in data between the data sets as well as within a dataset. Practically data of the same group of objects are usually stored in different datasets; in each data set, the data dimensions are not necessarily the same and unreal data may exist. Fuzzy clustering of a single data set would bring about less reliable results. And these data sets cannot be integrated for some reasons. An interesting application of vertical clustering occurs when dealing with huge data sets. Instead of clustering them in a single pass, we split them into individual data sets, cluster each of them separately, and reconcile the results through the collaborative exchange of prototypes. Vertical collaborative fuzzy C-Means is a useful tool for dealing collaborative clustering problems where a feature space is described in different pattern-sets. In this paper we use collaborative fuzzy clustering, first we cluster each data set individually and then optimize in accordance with the dependency of these datasets is adopted so as to improve the quality of fuzzy clustering of a single data set with the help of other data sets, taking personal privacy and security of data into consideration. © 2013 Springer-Verlag Berlin Heidelberg.
- Nanyang Technological University, Singapore
- Northumbria University, UK
- Federal University of Minas Gerais, Brazil
- Pontifical Catholic University of Chile, Chile
- National Chiao Tung University, Taiwan
- University of Auckland, New Zealand
- Mahanakorn University of Technology, Bangkok, Thailand
- Indian Institute of Technology Indore, India
- Jawaharlal Nehru University, New Delhi, India