I was a PhD Candidate and Postdoc Research Fellow in the School of Computer Science and Centre of Artificial Intelligence, UTS.
Currently, I am a Lecturer in the Discipline of ICT, University of Tasmania, Australia and appointed as an Adjunct Fellow in the School of Computer Science, UTS.
For more information, please go to my personal website.
Can supervise: YES
- Fuzzy Sets and Systems
- Fuzzy Neural Networks
- Deep Reinforcement Learning
- Game Artificial Intelligence
- Computational Neuroscience
- Brain-Computer Interface
- EEG / fNIR / fMRI Signal Processing
- Healthcare / Clinical Applications
Cao, Z, Ding, W, Wang, YK, Hussain, FK, Al-Jumaily, A & Lin, CT 2020, 'Effects of repetitive SSVEPs on EEG complexity using multiscale inherent fuzzy entropy', Neurocomputing, vol. 389, pp. 198-206.View/Download from: Publisher's site
© 2019 Elsevier B.V. Multiscale inherent fuzzy entropy is an objective measurement of electroencephalography (EEG) complexity, reflecting the habituation of brain systems. Entropy dynamics are generally believed to reflect the ability of the brain to adapt to a visual stimulus environment. In this study, we explored repetitive steady-state visual evoked potential (SSVEP)-based EEG complexity by assessing multiscale inherent fuzzy entropy with relative measurements. We used a wearable EEG device with Oz and Fpz electrodes to collect EEG signals from 40 participants under the following three conditions: a resting state (closed-eyes (CE) and open-eyes (OE) stimulation with five 15-Hz CE SSVEPs and stimulation with five 20-Hz OE SSVEPs. We noted monotonic enhancement of occipital EEG relative complexity with increasing stimulus times in CE and OE conditions. The occipital EEG relative complexity was significantly higher for the fifth SSVEP than for the first SSEVP (FDR-adjusted p < 0.05). Similarly, the prefrontal EEG relative complexity tended to be significantly higher in the OE condition compared to that in the CE condition (FDR-adjusted p < 0.05). The results also indicate that multiscale inherent fuzzy entropy is superior to other competing multiscale-based entropy methods. In conclusion, EEG relative complexity increases with stimulus times, a finding that reflects the strong habituation of brain systems. These results suggest that multiscale inherent fuzzy entropy is an EEG pattern with which brain complexity can be assessed using repetitive SSVEP stimuli.
Cao, Z, Lin, C-T, Lai, K-L, Ko, L-W, King, J-T, Liao, K-K, Fuh, J-L & Wang, S-J 2020, 'Extraction of SSVEPs-Based Inherent Fuzzy Entropy Using a Wearable Headband EEG in Migraine Patients', IEEE TRANSACTIONS ON FUZZY SYSTEMS, vol. 28, no. 1, pp. 14-27.View/Download from: Publisher's site
Cheng, C, Cao, Z & Xiao, F 2020, 'A generalized belief interval-valued soft set with applications in decision making', SOFT COMPUTING, vol. 24, no. 13, pp. 9339-9350.View/Download from: Publisher's site
IEEE As big data often contains a significant amount of uncertain, unstructured and imprecise data that are structurally complex and incomplete, traditional attribute reduction methods are less effective when applied to large-scale incomplete information systems to extract knowledge. Multigranular computing provides a powerful tool for use in big data analysis conducted at different levels of information granularity. In this paper, we present a novel multigranulation super-trust fuzzy-rough set-based attribute reduction (MSFAR) algorithm to support the formation of hierarchies of information granules of higher types and higher orders, which addresses newly emerging data mining problems in big data analysis. First, a multigranulation super-trust model based on the valued tolerance relation is constructed to identify the fuzzy similarity of the changing knowledge granularity with multimodality attributes. Second, an ensemble consensus compensatory scheme is adopted to calculate the multigranular trust degree based on the reputation at different granularities to create reasonable subproblems with different granulation levels. Third, an equilibrium method of multigranular-coevolution is employed to ensure a wide range of balancing of exploration and exploitation and can classify super elitists’ preferences and detect noncooperative behaviors with a global convergence ability and high search accuracy. The experimental results demonstrate that the MSFAR algorithm achieves a high performance in addressing uncertain and fuzzy attribute reduction problems with a large number of multigranularity variables.
Han, Y, Deng, Y, Cao, Z & Lin, CT 2020, 'An interval-valued Pythagorean prioritized operator-based game theoretical framework with its applications in multicriteria group decision making', Neural Computing and Applications.View/Download from: Publisher's site
© 2019, Springer-Verlag London Ltd., part of Springer Nature. Multicriteria decision-making process explicitly evaluates multiple conflicting criteria in decision making. The conventional decision-making approaches assumed that each agent is independent, but the reality is that each agent aims to maximize personal benefit which causes a negative influence on other agents' behaviors in a real-world competitive environment. In our study, we proposed an interval-valued Pythagorean prioritized operator-based game theoretical framework to mitigate the cross-influence problem. The proposed framework considers both prioritized levels among various criteria and decision makers within five stages. Notably, the interval-valued Pythagorean fuzzy sets are supposed to express the uncertainty of experts, and the game theories are applied to optimize the combination of strategies in interactive situations. Additionally, we also provided illustrative examples to address the application of our proposed framework. In summary, we provided a human-inspired framework to represent the behavior of group decision making in the interactive environment, which is potential to simulate the process of realistic humans thinking.
Hussain, T, Muhammad, K, Ullah, A, Cao, Z, Baik, SW & de Albuquerque, VHC 2020, 'Cloud-Assisted Multiview Video Summarization Using CNN and Bidirectional LSTM', IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, vol. 16, no. 1, pp. 77-86.View/Download from: Publisher's site
Li, F, Cao, Z, Xu, P, Yi, C, Liao, Y, Jiang, Y, Si, Y, Song, L, Zhang, T, Yao, D & Zhang, Y 2020, 'Reconfiguration of Brain Network between Rest-ing-state and P300 task', IEEE Transactions on Cognitive and Developmental Systems.View/Download from: Publisher's site
IEEE Previous studies explore the power spectra from the resting-state condition to the oddball task, but whether brain network existing significant difference is still unclear. Our study aims to address how the brain reconfigures its architecture from a resting-state condition (i.e., baseline) to the P300 task in the visual oddball task. In this study, electroencephalograms (EEGs) were collected from 24 subjects, who were required to only mentally count the number of target stimulus; afterwards, EEG networks constructed in different bands were compared between baseline and task to evaluate the reconfiguration of functional connectivity. Compared to the baseline, our results showed the significantly enhanced delta/theta functional connectivity and decreased alpha default mode network in the progress of brain reconfiguration to the task. Furthermore, the reconfigured coupling strengths were found to relate to P300 amplitudes, which were then regarded as features to train a classifier to differentiate the brain states and the high and low P300 groups with an accuracy of 100% and 77.78%, respectively. The findings of our study help us to under-stand the updates in functional connectivity from resting-state to the oddball task, and the reconfigured network structure has the potential for the selection of good subjects for P300-based brain-computer interface.
Li, Q, Cao, Z, Ding, W & Li, Q 2020, 'A multi-objective adaptive evolutionary algorithm to extract communities in networks', SWARM AND EVOLUTIONARY COMPUTATION, vol. 52.View/Download from: Publisher's site
Li, Q, Cao, Z, Tanveer, M, Pandey, HM & Wang, C 2020, 'A Semantic Collaboration Method Based on Uniform Knowledge Graph', IEEE INTERNET OF THINGS JOURNAL, vol. 7, no. 5, pp. 4473-4484.View/Download from: Publisher's site
Li, Q, Zhong, J, Cao, Z & Li, X 2020, 'Optimizing streaming graph partitioning via a heuristic greedy method and caching strategy', Optimization Methods and Software.View/Download from: Publisher's site
© 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group. Graph partitioning is an important method for accelerating large distributed graph computation. Streaming graph partitioning is more efficient than offline partitioning, and it has been developed continuously in the application of graph partitioning in recent years. In this work, we first introduce a heuristic greedy streaming partitioning method and show that it outperforms the state-of-the-art streaming partitioning methods, leading to exact balance and fewer cut edges. Second, we propose a cache structure for streaming partitioning, called an adjacent edge structure, which can improve the partition efficiency several times on a single commodity type computer without affecting the partition quality. Regardless as to whether the memory capacity is limited (local cache) or not (global cache), our strategy can also improve the partition quality by restreaming partitioning. Taking linear weight greedy streaming algorithm as an example, the experimental results on 19 real-world graphs show that the average partitioning time of the new method is 4.9 times faster than that of the original method, which proves the effectiveness and superiority of the cache structure mentioned in this paper.
Liu, Z, Xiao, F, Lin, C-T, Kang, BH & Cao, Z 2020, 'A Generalized Golden Rule Representative Value for Multiple-Criteria Decision Analysis', IEEE Transactions on Systems, Man, and Cybernetics: Systems, pp. 1-12.View/Download from: Publisher's site
Si, Y, Li, F, Duan, K, Tao, Q, Li, C, Cao, Z, Zhang, Y, Biswal, B, Li, P, Yao, D & Xu, P 2020, 'Predicting individual decision-making responses based on single-trial EEG.', NeuroImage, vol. 206, pp. 116333-116333.View/Download from: Publisher's site
Decision-making plays an essential role in the interpersonal interactions and cognitive processing of individuals. There has been increasing interest in being able to predict an individual's decision-making response (i.e., acceptance or rejection). We proposed an electroencephalogram (EEG)-based computational intelligence framework to predict individual responses. Specifically, the discriminative spatial network pattern (DSNP), a supervised learning approach, was applied to single-trial EEG data to extract the DSNP feature from the single-trial brain network. A linear discriminate analysis (LDA) trained on the DSNP features was then used to predict the individual response trial-by-trial. To verify the performance of the proposed DSNP, we recruited two independent subject groups, and recorded the EEGs using two types of EEG systems. The performances of the trial-by-trial predictors achieved an accuracy of 0.88 ± 0.09 for the first dataset, and 0.90 ± 0.10 for the second dataset. These trial-by-trial prediction performances suggested that individual responses could be predicted trial-by-trial by using the specific pattern of single-trial EEG networks, and our proposed method has the potential to establish the biologically inspired artificial intelligence decision system.
Wang, Q, Zhou, Y, Ding, W, Zhang, Z, Muhammad, K & Cao, Z 2020, 'Random forest with self-paced bootstrap learning in lung cancer prognosis', ACM Transactions on Multimedia Computing, Communications and Applications, vol. 16, no. 1s.View/Download from: Publisher's site
© 2020 ACM. Training gene expression data with supervised learning approaches can provide an alarm sign for early treatment of lung cancer to decrease death rates. However, the samples of gene features involve lots of noises in a realistic environment. In this study, we present a random forest with self-paced learning bootstrap for improvement of lung cancer classification and prognosis based on gene expression data. To be specific, we propose an ensemble learning with random forest approach to improving the model classification performance by selecting multi-classifiers. Then, we investigate the sampling strategy by gradually embedding from high- to low-quality samples by self-paced learning. The experimental results based on five public lung cancer datasets show that our proposed method could select significant genes exactly, which improves classification performance compared to that of existing approaches. We believe that our proposed method has the potential to assist doctors in gene selections and lung cancer prognosis.
Yang, Z, Garg, H, Li, J, Srivastava, G & Cao, Z 2020, 'Investigation of multiple heterogeneous relationships using a q-rung orthopair fuzzy multi-criteria decision algorithm', NEURAL COMPUTING & APPLICATIONS.View/Download from: Publisher's site
We describe driver behaviour and brain dynamics acquired from a 90-minute sustained-attention task in an immersive driving simulator. The data included 62 sessions of 32-channel electroencephalography (EEG) data for 27 subjects driving on a four-lane highway who were instructed to keep the car cruising in the centre of the lane. Lane-departure events were randomly induced to cause the car to drift from the original cruising lane towards the left or right lane. A complete trial included events with deviation onset, response onset, and response offset. The next trial, in which the subject was instructed to drive back to the original cruising lane, began 5-10 seconds after finishing the previous trial. We believe that this dataset will lead to the development of novel neural processing methodology that can be used to index brain cortical dynamics and detect driving fatigue and drowsiness. This publicly available dataset will be beneficial to the neuroscience and brain-computer interface communities.
Cao, Z, Lin, CT, Ding, W, Chen, MH, Li, CT & Su, TP 2019, 'Identifying Ketamine Responses in Treatment-Resistant Depression Using a Wearable Forehead EEG', IEEE Transactions on Biomedical Engineering, vol. 66, no. 6, pp. 1668-1679.View/Download from: Publisher's site
IEEE This study explores the responses to ketamine in patients with treatment-resistant depression (TRD) using a wearable forehead electroencephalography (EEG) device. We recruited 55 outpatients with TRD who were randomized into three approximately equal- sized groups (A: 0.5 mg/kg ketamine; B: 0.2 mg/kg ketamine; and C: normal saline) under double-blind conditions. The ketamine responses were measured by EEG signals and Hamilton Depression Rating Scale (HDRS) scores. At baseline, responders showed a significantly weaker EEG theta power than did non- responders (p < 0.05). Responders exhibited a higher EEG alpha power but lower EEG alpha asymmetry and theta cordance at post-treatment than at baseline (p < 0.05). Furthermore, our baseline EEG predictor classified responders and non-responders with 81.3 $\pm$ 9.5% accuracy, 82.1 $\pm$ 8.6% sensitivity and 91.9 $\pm$ 7.4% specificity. In conclusion, the rapid antidepressant effects of mixed doses of ketamine are associated with prefrontal EEG power, asymmetry and cordance at baseline and early post-treatment changes. The prefrontal EEG patterns at baseline may account for recognizing ketamine effects in advance. Our randomized, double- blind, placebo-controlled study provides information regarding clinical impacts on the potential targets underlying baseline identification and early changes from the effects of ketamine in patients with TRD.
Chen, S, Wang, Y, Lin, CT, Ding, W & Cao, Z 2019, 'Semi-supervised feature learning for improving writer identification', Information Sciences, vol. 482, pp. 156-170.View/Download from: Publisher's site
© 2019 Elsevier Inc. Data augmentation is typically used by supervised feature learning approaches for offline writer identification, but such approaches require a mass of additional training data and potentially lead to overfitting errors. In this study, a semi-supervised feature learning pipeline is proposed to improve the performance of writer identification by training with extra unlabeled data and the original labeled data simultaneously. Specifically, we propose a weighted label smoothing regularization (WLSR) method for data augmentation, which assigns a weighted uniform label distribution to the extra unlabeled data. The WLSR method regularizes the convolutional neural network (CNN) baseline to allow more discriminative features to be learned to represent the properties of different writing styles. The experimental results on well-known benchmark datasets (ICDAR2013 and CVL) showed that our proposed semi-supervised feature learning approach significantly improves the baseline measurement and perform competitively with existing writer identification approaches. Our findings provide new insights into offline writer identification.
Ding, W, Lin, C-T & Cao, Z 2019, 'Deep Neuro-Cognitive Co-Evolution for Fuzzy Attribute Reduction by Quantum Leaping PSO With Nearest-Neighbor Memeplexes.', IEEE Transactions on Cybernetics, vol. 49, no. 7, pp. 2744-2757.View/Download from: Publisher's site
Attribute reduction with many patterns and indicators has been regarded as an important approach for large-scale data mining and machine learning tasks. However, it is extremely difficult for researchers to inadequately extract knowledge and insights from multiple overlapping and interdependent fuzzy datasets from the current changing and interconnected big data sources. This paper proposes a deep neuro-cognitive co-evolution for fuzzy attribute reduction (DNCFAR) that contains a combination of quantum leaping particle swarm optimization with nearest-neighbor memeplexes. A key element of DNCFAR resides in its deep neuro-cognitive cooperative co-evolution structure, which is explicitly permitted to identify interdependent variables and adaptively decompose them in the same neuro-subpopulation, with minimizing the complexity and nonseparability of interdependent variables among different fuzzy attribute subsets. Next DNCFAR formalizes to the different types of quantum leaping particles with nearest-neighbor memeplexes to share their respective solutions and deeply cooperate to evolve the assigned fuzzy attribute subsets. The experimental results demonstrate that DNCFAR can achieve competitive performance in terms of average computational efficiency and classification accuracy while reinforcing noise tolerance. Furthermore, it can be well applied to clearly identify different longitudinal surfaces of infant cerebrum regions, which indicates its great potential for brain disorder prediction based on fMRI.
Ding, W, Lin, C-T & Cao, Z 2019, 'Shared Nearest-Neighbor Quantum Game-Based Attribute Reduction With Hierarchical Coevolutionary Spark and Its Application in Consistent Segmentation of Neonatal Cerebral Cortical Surfaces.', IEEE Transactions on Neural Networks and Learning Systems, vol. 30, no. 7, pp. 2013-2027.View/Download from: Publisher's site
The unprecedented increase in data volume has become a severe challenge for conventional patterns of data mining and learning systems tasked with handling big data. The recently introduced Spark platform is a new processing method for big data analysis and related learning systems, which has attracted increasing attention from both the scientific community and industry. In this paper, we propose a shared nearest-neighbor quantum game-based attribute reduction (SNNQGAR) algorithm that incorporates the hierarchical coevolutionary Spark model. We first present a shared coevolutionary nearest-neighbor hierarchy with self-evolving compensation that considers the features of nearest-neighborhood attribute subsets and calculates the similarity between attribute subsets according to the shared neighbor information of attribute sample points. We then present a novel attribute weight tensor model to generate ranking vectors of attributes and apply them to balance the relative contributions of different neighborhood attribute subsets. To optimize the model, we propose an embedded quantum equilibrium game paradigm (QEGP) to ensure that noisy attributes do not degrade the big data reduction results. A combination of the hierarchical coevolutionary Spark model and an improved MapReduce framework is then constructed that it can better parallelize the SNNQGAR to efficiently determine the preferred reduction solutions of the distributed attribute subsets. The experimental comparisons demonstrate the superior performance of the SNNQGAR, which outperforms most of the state-of-the-art attribute reduction algorithms. Moreover, the results indicate that the SNNQGAR can be successfully applied to segment overlapping and interdependent fuzzy cerebral tissues, and it exhibits a stable and consistent segmentation performance for neonatal cerebral cortical surfaces.
© 2019 Elsevier B.V. Network embedding aims at learning the low dimensional representation of nodes. These representations can be widely used for network mining tasks, such as link prediction, anomaly detection, and classification. Recently, a great deal of meaningful research work has been carried out on this emerging network analysis paradigm. The real-world network contains different size clusters because of the edges with different relationship types. These clusters also reflect some features of nodes, which can contribute to the optimization of the feature representation of nodes. However, existing network embedding methods do not distinguish these relationship types. In this paper, we propose an unsupervised network representation learning model that can encode edge relationship information. Firstly, an objective function is defined, which can learn the edge vectors by implicit clustering. Then, a biased random walk is designed to generate a series of node sequences, which are put into Skip-Gram to learn the low dimensional node representations. Extensive experiments are conducted on several network datasets. Compared with the state-of-art baselines, the proposed method is able to achieve favorable and stable results in multi-label classification and link prediction tasks.
Li, Q, Zhong, J, Li, Q, Wang, C & Cao, Z 2019, 'A community merger of optimization algorithm to extract overlapping communities in networks', IEEE Access, vol. 7, pp. 3994-4005.View/Download from: Publisher's site
© 2018 IEEE. A community in networks is a subset of vertices primarily connecting internal components, yet less connecting to the external vertices. The existing algorithms aim to extract communities of the topological features in networks. However, the edges of practical complex networks involving a weight that represents the tightness degree of connection and robustness, which leads a significant influence on the accuracy of community detection. In our study, we propose an overlapping community detection method based on the seed expansion strategy applying to both the unweighted and the weighted networks, called OCSE. First, it redefines the edge weight and the vertex weight depending on the influence of the network topology and the original edge weight, and then selects the seed vertices and updates the edges weight. Comparisons between OCSE approach and existing community detection methods on synthetic and real-world networks, the results of the experiment show that our proposed approach has the significantly better performance in terms of the accuracy.
Ma, W, Cai, L, He, T, Chen, L, Cao, Z & Li, R 2019, 'Local Expansion and Optimization for Higher-Order Graph Clustering', IEEE Internet of Things Journal, vol. 6, no. 5, pp. 8702-8713.View/Download from: Publisher's site
© 2014 IEEE. Graph clustering aims to identify clusters that feature tighter connections between internal nodes than external nodes. We noted that conventional clustering approaches based on a single vertex or edge cannot meet the requirements of clustering in a higher-order mixed structure formed by multiple nodes in a complex network. Considering the above limitation, we are aware of the fact that a clustering coefficient can measure the degree to which nodes in a graph tend to cluster, even if only a small area of the graph is given. In this paper, we introduce a new cluster quality score, i.e., the local motif rate, which can effectively respond to the density of clusters in a higher-order graph. We also propose a motif-based local expansion and optimization algorithm (MLEO) to improve local higher-order graph clustering. This algorithm is a purely local algorithm and can be applied directly to higher-order graphs without conversion to a weighted graph, thus avoiding distortion of the transform. In addition, we propose a new seed-processing strategy in a higher-order graph. The experimental results show that our proposed strategy can achieve better performance than the existing approaches when using a quadrangle as the motif in the LFR network and the value of the mixing parameter \mu exceeds 0.6.
Xia, L-Y, Wang, Q-Y, Cao, Z & Liang, Y 2019, 'Descriptor Selection Improvements for Quantitative Structure-Activity Relationships', INTERNATIONAL JOURNAL OF NEURAL SYSTEMS, vol. 29, no. 9.View/Download from: Publisher's site
Yang, Z, Li, X, Cao, Z & Li, J 2019, 'Q-rung Orthopair Normal Fuzzy Aggregation Operators and Their Application in Multi-Attribute Decision-Making', MATHEMATICS, vol. 7, no. 12.View/Download from: Publisher's site
Yang, Z, Xiong, G, Cao, Z, Li, Y & Huang, L 2019, 'A Decision Method for Online Purchases Considering Dynamic Information Preference Based on Sentiment Orientation Classification and Discrete DIFWA Operators', IEEE ACCESS, vol. 7, pp. 77008-77026.View/Download from: Publisher's site
Brown, P, RELISH Consortium & Zhou, Y 2019, 'Large expert-curated database for benchmarking document similarity in biomedical literature search', Database: the journal of biological databases and curation, vol. 2019, pp. 1-66.View/Download from: Publisher's site
Document recommendation systems for locating relevant literature have mostly relied on methods developed a decade ago. This is largely due to the lack of a large offline gold-standard benchmark of relevant documents that cover a variety of research fields such that newly developed literature search techniques can be compared, improved and translated into practice. To overcome this bottleneck, we have established the RElevant LIterature SearcH consortium consisting of more than 1500 scientists from 84 countries, who have collectively annotated the relevance of over 180 000 PubMed-listed articles with regard to their respective seed (input) article/s. The majority of annotations were contributed by highly experienced, original authors of the seed articles. The collected data cover 76% of all unique PubMed Medical Subject Headings descriptors. No systematic biases were observed across different experience levels, research fields or time spent on annotations. More importantly, annotations of the same document pairs contributed by different scientists were highly concordant. We further show that the three representative baseline methods used to generate recommended articles for evaluation (Okapi Best Matching 25, Term Frequency–Inverse Document Frequency and PubMed Related Articles) had similar overall performances. Additionally, we found that these methods each tend to produce distinct collections of recommended articles, suggesting that a hybrid method may be required to completely capture all relevant articles. The established database server located at https://relishdb.ict.griffith.edu.au is freely available for the downloading of annotation data and the blind testing of new methods. We expect that this benchmark will be useful for stimulating the development of new powerful techniques for title and title/abstract-based search engines for relevant articles in biomedical research.
Cao, Z, Lai, K-L, Lin, C-T, Chuang, C-H, Chou, C-C & Wang, S-J 2018, 'Exploring resting-state EEG complexity before migraine attacks.', Cephalalgia: an international journal of headache, vol. 38, no. 7, pp. 1296-1306.View/Download from: Publisher's site
Objective Entropy-based approaches to understanding the temporal dynamics of complexity have revealed novel insights into various brain activities. Herein, electroencephalogram complexity before migraine attacks was examined using an inherent fuzzy entropy approach, allowing the development of an electroencephalogram-based classification model to recognize the difference between interictal and preictal phases. Methods Forty patients with migraine without aura and 40 age-matched normal control subjects were recruited, and the resting-state electroencephalogram signals of their prefrontal and occipital areas were prospectively collected. The migraine phases were defined based on the headache diary, and the preictal phase was defined as within 72 hours before a migraine attack. Results The electroencephalogram complexity of patients in the preictal phase, which resembled that of normal control subjects, was significantly higher than that of patients in the interictal phase in the prefrontal area (FDR-adjusted p < 0.05) but not in the occipital area. The measurement of test-retest reliability (n = 8) using the intra-class correlation coefficient was good with r1 = 0.73 ( p = 0.01). Furthermore, the classification model, support vector machine, showed the highest accuracy (76 ± 4%) for classifying interictal and preictal phases using the prefrontal electroencephalogram complexity. Conclusion Entropy-based analytical methods identified enhancement or "normalization" of frontal electroencephalogram complexity during the preictal phase compared with the interictal phase. This classification model, using this complexity feature, may have the potential to provide a preictal alert to migraine without aura patients.
Chuang, C-H, Cao, Z, King, J-T, Wu, B-S, Wang, Y-K & Lin, C-T 2018, 'Brain Electrodynamic and Hemodynamic Signatures Against Fatigue During Driving.', Frontiers in Neuroscience, vol. 12, pp. 1-12.View/Download from: Publisher's site
Fatigue is likely to be gradually cumulated in a prolonged and attention-demanding task that may adversely affect task performance. To address the brain dynamics during a driving task, this study recruited 16 subjects to participate in an event-related lane-departure driving experiment. Each subject was instructed to maintain attention and task performance throughout an hour-long driving experiment. The subjects' brain electrodynamics and hemodynamics were simultaneously recorded via 32-channel electroencephalography (EEG) and 8-source/16-detector functional near-infrared spectroscopy (fNIRS). The behavior performance demonstrated that all subjects were able to promptly respond to lane-deviation events, even if the sign of fatigue arose in the brain, which suggests that the subjects were fighting fatigue during the driving experiment. The EEG event-related analysis showed strengthening alpha suppression in the occipital cortex, a common brain region of fatigue. Furthermore, we noted increasing oxygenated hemoglobin (HbO) of the brain to fight driving fatigue in the frontal cortex, primary motor cortex, parieto-occipital cortex and supplementary motor area. In conclusion, the increasing neural activity and cortical activations were aimed at maintaining driving performance when fatigue emerged. The electrodynamic and hemodynamic signatures of fatigue fighting contribute to our understanding of the brain dynamics of driving fatigue and address driving safety issues through the maintenance of attention and behavioral performance.
Cao, Z & Lin, CT 2018, 'Inherent Fuzzy Entropy for the Improvement of EEG Complexity Evaluation', IEEE Transactions on Fuzzy Systems, vol. 26, no. 2, pp. 1032-1035.View/Download from: Publisher's site
© 2017 IEEE. In recent years, the concept of entropy has been widely used to measure the dynamic complexity of signals. Since the state of complexity of human beings is significantly affected by their health state, developing accurate complexity evaluation algorithms is a crucial and urgent area of study. This paper proposes using inherent fuzzy entropy (Inherent FuzzyEn) and its multiscale version, which employs empirical mode decomposition and fuzzy membership function (exponential function) to address the dynamic complexity in electroencephalogram (EEG) data. In the literature, the reliability of entropy-based complexity evaluations has been limited by superimposed trends in signals and a lack of multiple time scales. Our proposed method represents the first attempt to use the Inherent FuzzyEn algorithm to increase the reliability of complexity evaluation in realistic EEG applications. We recorded the EEG signals of several subjects under resting condition, and the EEG complexity was evaluated using approximate entropy, sample entropy, FuzzyEn, and Inherent FuzzyEn, respectively. The results indicate that Inherent FuzzyEn is superior to other competing models regardless of the use of fuzzy or nonfuzzy structures, and has the most stable complexity and smallest root mean square deviation.
OAPA Fusing multiple existing models for filtering webpages can mitigate the shortcomings of individual filtering models. To provide an engine for such fusion, we propose a multimodel fusion engine for filtering webpages (MMFEFWP) for the extraction of target webpages. This engine can handle large datasets of webpages crawled from websites and supports five individual filtering models and the fusion of any two of them. There are two possible fusion methods: one is to simultaneously satisfy the conditions of both individual models, and the other is to satisfy the conditions of one of the two individual models. We present the functions, architecture, and software design of the proposed engine. We use recall ratio (RR) and precision ratio (PR) as the evaluation indices of the filtering models and propose rules describing how PR and RR change when individual models are fused. We use 200,000 webpages collected by crawling the popular online shopping website "www.jd.com" as the experimental dataset to verify these rules. The experimental results show that two-model fusion can improve either PR or RR. Thus, the proposed engine has good practical value for engineering applications.
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: 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.
He, T, Cai, L, Meng, T, Chen, L, Deng, Z & Cao, Z 2018, 'Parallel Community Detection Based on Distance Dynamics for Large-Scale Network', IEEE Access, vol. 6, pp. 42775-42789.View/Download from: Publisher's site
© 2013 IEEE. Data mining task is a challenge on finding a high-quality community structure from large-scale networks. The distance dynamics model was proved to be active on regular-size network community, but it is difficult to discover the community structure effectively from the large-scale network (0.1-1 billion edges), due to the limit of machine hardware and high time complexity. In this paper, we proposed a parallel community detection algorithm based on the distance dynamics model called P-Attractor, which is capable of handling the detection problem of large networks community. Our algorithm first developed a graph partitioning method to divide large network into lots of sub-networks, yet maintaining the complete neighbor structure of the original network. Then, the traditional distance dynamics model was improved by the dynamic interaction process to simulate the distance evolution of each sub-network. Finally, we discovered the real community structure by removing all external edges after evolution process. In our extensive experiments on multiple synthetic networks and real-world networks, the results showed the effectiveness and efficiency of P-Attractor, and the execution time on 4 threads and 32 threads are around 10 and 2 h, respectively. Our proposed algorithm is potential to discover community from a billion-scale network, such as Uk-2007.
Meng, T, Cai, L, He, T, Chen, L, Deng, Z, Ding, W & Cao, Z 2018, 'A Modified Distance Dynamics Model for Improvement of Community Detection', IEEE Access, vol. 6, pp. 63934-63947.View/Download from: Publisher's site
© 2018 IEEE. Community detection is a key technique for identifying the intrinsic community structures of complex networks. The distance dynamics model has been proven effective in finding communities with arbitrary size and shape and identifying outliers. However, to simulate distance dynamics, the model requires manual parameter specification and is sensitive to the cohesion threshold parameter, which is difficult to determine. Furthermore, it has difficulty handling rough outliers and ignores hubs (nodes that bridge communities). In this paper, we propose a robust distance dynamics model, namely, Attractor++, which uses a dynamic membership degree. In Attractor++, the dynamic membership degree is used to determine the influence of exclusive neighbors on the distance instead of setting the cohesion threshold. In addition, considering its inefficiency and low accuracy in handling outliers and identifying hubs, we design an outlier optimization model that is based on triangle adjacency. By using optimization rules, a postprocessing method further judges whether a singleton node should be merged into the same community as its triangles or regarded as a hub or an outlier. Extensive experiments on both real-world and synthetic networks demonstrate that our algorithm more accurately identifies nodes that have special roles (hubs and outliers) and more effectively identifies community structures.
Lin, C, Liu, Y-T, Wu, S-L, Cao, Z, Wang, Y, Huang, C-S, King, J-T, Chen, S-A, Lu, S-W & Chuang, C 2017, 'EEG-Based Brain-Computer Interfaces: A Novel Neurotechnology and Computational Intelligence Method', IEEE Systems Man and Cybernetics Magazine, vol. 3, no. 4, pp. 16-26.View/Download from: Publisher's site
This article presents the latest BCI-related research done in our group. Our previous work applied computational intelligence technology in BCIs to inspire detailed investigations of practical issues in real-life applications. Novel EEG devices featuring dry electrodes facilitate and speed up electrode positioning before recording and allow subjects to move freely in operational environments. We also demonstrate the feasibility of applying CCA, RBFNs, effective connectivity measurements, and D-S theory to help BCIs extract informative knowledge from brain signals. Two recent trends in research in the computational and artificial intelligence community, big data and deep learning, are expected to impact the direction and development of BCIs.
Lin, CT, Chuang, CH, Cao, Z, Singh, AK, Hung, CS, Yu, YH, Nascimben, M, Liu, YT, King, JT, Su, TP & Wang, SJ 2017, 'Forehead EEG in Support of Future Feasible Personal Healthcare Solutions: Sleep Management, Headache Prevention, and Depression Treatment', IEEE Access, vol. 5, pp. 10612-10621.View/Download from: Publisher's site
© 2013 IEEE. There are current limitations in the recording technologies for measuring EEG activity in clinical and experimental applications. Acquisition systems involving wet electrodes are time-consuming and uncomfortable for the user. Furthermore, dehydration of the gel affects the quality of the acquired data and reliability of long-term monitoring. As a result, dry electrodes may be used to facilitate the transition from neuroscience research or clinical practice to real-life applications. EEG signals can be easily obtained using dry electrodes on the forehead, which provides extensive information concerning various cognitive dysfunctions and disorders. This paper presents the usefulness of the forehead EEG with advanced sensing technology and signal processing algorithms to support people with healthcare needs, such as monitoring sleep, predicting headaches, and treating depression. The proposed system for evaluating sleep quality is capable of identifying five sleep stages to track nightly sleep patterns. Additionally, people with episodic migraines can be notified of an imminent migraine headache hours in advance through monitoring forehead EEG dynamics. The depression treatment screening system can predict the efficacy of rapid antidepressant agents. It is evident that frontal EEG activity is critically involved in sleep management, headache prevention, and depression treatment. The use of dry electrodes on the forehead allows for easy and rapid monitoring on an everyday basis. The advances in EEG recording and analysis ensure a promising future in support of personal healthcare solutions.
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, IEEE International Conference on Systems, Man and Cybernetics, IEEE, Bari, Italy, 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.
Li, Q, Zhong, J, Li, Q, Cao, Z & Wang, C 2019, 'Enhancing network embedding with implicit clustering', Database Systems for Advanced Applications (LNCS), International Conference on Database Systems for Advanced Applications, Springer, Chiang Mai, Thailand, pp. 452-467.View/Download from: Publisher's site
© Springer Nature Switzerland AG 2019. Network embedding aims at learning the low dimensional representation of nodes. These representations can be widely used for network mining tasks, such as link prediction, anomaly detection, and classification. Recently, a great deal of meaningful research work has been carried out on this emerging network analysis paradigm. The real-world network contains different size clusters because of the edges with different relationship types. These clusters also reflect some features of nodes, which can contribute to the optimization of the feature representation of nodes. However, existing network embedding methods do not distinguish these relationship types. In this paper, we propose an unsupervised network representation learning model that can encode edge relationship information. Firstly, an objective function is defined, which can learn the edge vectors by implicit clustering. Then, a biased random walk is designed to generate a series of node sequences, which are put into Skip-Gram to learn the low dimensional node representations. Extensive experiments are conducted on several network datasets. Compared with the state-of-art baselines, the proposed method is able to achieve favorable and stable results in multi-label classification and link prediction tasks.
Park, M, Yang, W, Cao, Z, Kang, B, Connor, D & Lea, MA 2019, 'Marine Vertebrate Predator Detection and Recognition in Underwater Videos by Region Convolutional Neural Network', Knowledge Management and Acquisition for Intelligent Systems, Pacific Rim Knowledge Acquisition Workshop, Springer, Cuvu, Fiji, pp. 66-80.View/Download from: Publisher's site
© 2019, Springer Nature Switzerland AG. In this paper, we present R-CNN, Fast R-CNN and Faster R-CNN methods to automatically detect and recognise the predators in underwater videos. We compare the results of these methods on real data and discuss their strengths and weaknesses. We build a dataset using footage captured from representative environment of the wild and devise a data model with three classes (seal, dolphin, background). Following this, we train R-CNN, Fast R-CNN and Faster R-CNN, then evaluate them on a test dataset compose of challenging objects that had not been seen during training. We perform evaluation on GPU, acquiring information about the AP and IOU for each model and network based on various proposal numbers as well as runtime speeds. Based on the results, we found that the best model of predator detection using visual deep learning models is Faster R-CNN with 2000 proposals.
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: 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.
Cao, ZH, Ko, LW, Lai, KL, Huang, SB, Wang, SJ & Lin, CT 2015, 'Classification of migraine stages based on resting-state EEG power', Proceedings of the International Joint Conference on Neural Networks, IEEE International Joint Conference on Neural Networks, IEEE, Killarney, Ireland.View/Download from: Publisher's site
© 2015 IEEE. Migraine is a chronic neurological disease characterized by recurrent moderate to severe headaches during a period like one month often in association with symptoms in human brain and autonomic nervous system. Normally, migraine symptoms can be categorized into four different stages: inter-ictal, pre-ictal, ictal, and post-ictal stages. Since migraine patients are difficulty knowing when they will suffer migraine attacks, therefore, early detection becomes an important issue, especially for low-frequency migraine patients who have less than 5 times attacks per month. The main goal of this study is to develop a migraine-stage classification system based on migraineurs' resting-state EEG power. We collect migraineurs' O1 and O2 EEG activities during closing eyes from occipital lobe to identify pre-ictal and non-pre-ictal stages. Self-Constructing Neural Fuzzy Inference Network (SONFIN) is adopted as the classifier in the migraine stages classification which can reach the better classification accuracy (66%) in comparison with other classifiers. The proposed system is helpful for migraineurs to obtain better treatment at the right time.