Dr Junyu Xuan is ARC Discovery Early Career Researcher Award (DECRA) and Lecturer of Australia Artificial Intelligence Institue in Faculty of Engineering and IT at the University of Technology Sydney (UTS).
Dr Xuan received Dual-doctoral degree from University of Technology Sydney and Shanghai University in 2016. After PhD, he worked as a Postdoctoral Research Fellow in Faculty of Engineering and IT of the University of Technology Sydney in Australia for three years. In 2020, he secured an ARC Discovery Early Career Researcher Award and became a Lecturer at UTS.
Dr Xuan’s research interests include Machine Learning, Bayesian Nonparametric Learning, Text Mining, Web Mining, etc. He has published 40 papers in high-quality journals and conferences, including Artificial Intelligence Journal, Machine Learning Journal, IEEE TNNLS, IEEE TFS, IEEE TKDE, IEEE TCYB, ACM TOIS, ACM TIST, ACM Computing Surveys, ICDM, NeurIPS, AAAI, IJCNN, etc.
- Member of International Society for Bayesian Analysis (ISBA)
- Member of Association for Computing Machinery (ACM)
- Member of Electrical and Electronics Engineers (IEEE)
Can supervise: YES
- Bayesian Nonparametric Learning
- Bayesian Deep Learning
- Probabilistic Graphical Model
- Text Mining
- Web Mining
Xuan, J, Luo, X, Lu, J & Zhang, G 2020, 'Web event evolution trend prediction based on its computational social context', WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS, vol. 23, no. 3, pp. 1861-1886.View/Download from: Publisher's site
Lin, A, Lu, J, Xuan, J, Zhu, F & Zhang, G 2020, 'A Causal Dirichlet Mixture Model for Causal Inference from Observational Data', ACM Transactions on Intelligent Systems and Technology, vol. 11, no. 3.View/Download from: Publisher's site
© 2020 ACM. Estimating causal effects by making causal inferences from observational data is common practice in scientific studies, business decision-making, and daily life. In today's data-driven world, causal inference has become a key part of the evaluation process for many purposes, such as examining the effects of medicine or the impact of an economic policy on society. However, although the literature contains some excellent models, there is room to improve their representation power and their ability to capture complex relationships. For these reasons, we propose a novel prior called Causal DP and a model called CDP. The prior captures the complex relationships between covariates, treatments, and outcomes in observational data using a rational probabilistic dependency structure. The model is Bayesian, nonparametric, and generative and is not based on the assumption of any parametric distribution. CDP is designed to estimate various kinds of causal effects - average, conditional average, average treated, quantile, and so on. It performs well with missing covariates and does not suffer from overfitting. Comparative experiments on synthetic datasets against several state-of-the-art methods demonstrate that CDP has a superior ability to capture complex relationships. Further, a simple evaluation to infer the effect of a job training program on trainee earnings from real-world data shows that CDP is both effective and useful for causal inference.
Liu, Q, Huang, H, Xuan, J, Zhang, G, Gao, Y & Lu, J 2020, 'A Fuzzy Word Similarity Measure for Selecting Top-k Similar Words in Query Expansion', IEEE Transactions on Fuzzy Systems, pp. 1-1.View/Download from: Publisher's site
The cooperative hierarchical structure is a common and significant data
structure observed in, or adopted by, many research areas, such as: text mining
(author-paper-word) and multi-label classification (label-instance-feature).
Renowned Bayesian approaches for cooperative hierarchical structure modeling
are mostly based on topic models. However, these approaches suffer from a
serious issue in that the number of hidden topics/factors needs to be fixed in
advance and an inappropriate number may lead to overfitting or underfitting.
One elegant way to resolve this issue is Bayesian nonparametric learning, but
existing work in this area still cannot be applied to cooperative hierarchical
In this paper, we propose a cooperative hierarchical Dirichlet process (CHDP)
to fill this gap. Each node in a cooperative hierarchical structure is assigned
a Dirichlet process to model its weights on the infinite hidden factors/topics.
Together with measure inheritance from hierarchical Dirichlet process, two
kinds of measure cooperation, i.e., superposition and maximization, are defined
to capture the many-to-many relationships in the cooperative hierarchical
structure. Furthermore, two constructive representations for CHDP, i.e.,
stick-breaking and international restaurant process, are designed to facilitate
the model inference. Experiments on synthetic and real-world data with
cooperative hierarchical structures demonstrate the properties and the ability
of CHDP for cooperative hierarchical structure modeling and its potential for
practical application scenarios.
Lu, J, Xuan, J, Zhang, G & Luo, X 2018, 'Structural property-aware multilayer network embedding for latent factor analysis', Pattern Recognition, vol. 76, pp. 228-241.View/Download from: Publisher's site
© 2017 Elsevier Ltd Multilayer network is a structure commonly used to describe and model the complex interaction between sets of entities/nodes. A three-layer example is the author-paper-word structure in which authors are linked by co-author relation, papers are linked by citation relation, and words are linked by semantic relation. Network embedding, which aims to project the nodes in the network into a relatively low-dimensional space for latent factor analysis, has recently emerged as an effective method for a variety of network-based tasks, such as collaborative filtering and link prediction. However, existing studies of network embedding both focus on the single-layer network and overlook the structural properties of the network, e.g., the degree distribution and communities, which are significant for node characterization, such as the preferences of users in a social network. In this paper, we propose four multilayer network embedding algorithms based on Nonnegative Matrix Factorization (NMF) with consideration given to four structural properties: whole network (NNMF), community (CNMF), degree distribution (DNMF), and max spanning tree (TNMF). Experiments on synthetic data show that the proposed algorithms are able to preserve the desired structural properties as designed. Experiments on real-world data show that multilayer network embedding improves the accuracy of document clustering and recommendation, and the four embedding algorithms corresponding to the four structural properties demonstrate the differences in performance on these two tasks. These results can be directly used in document clustering and recommendation systems.
Xuan, J, Lu, J, Zhang, G, Xu, RYD & Luo, X 2018, 'Doubly Nonparametric Sparse Nonnegative Matrix Factorization Based on Dependent Indian Buffet Processes', IEEE Transactions on Neural Networks and Learning Systems, pp. 1835-1849.View/Download from: Publisher's site
Sparse nonnegative matrix factorization (SNMF) aims to factorize a data matrix into two optimized nonnegative sparse factor matrices, which could benefit many tasks, such as document-word co-clustering. However, the traditional SNMF typically assumes the number of latent factors (i.e., dimensionality of the factor matrices) to be fixed. This assumption makes it inflexible in practice. In this paper, we propose a doubly sparse nonparametric NMF framework to mitigate this issue by using dependent Indian buffet processes (dIBP). We apply a correlation function for the generation of two stick weights associated with each column pair of factor matrices while still maintaining their respective marginal distribution specified by IBP. As a consequence, the generation of two factor matrices will be columnwise correlated. Under this framework, two classes of correlation function are proposed: 1) using bivariate Beta distribution and 2) using Copula function. Compared with the single IBP-based NMF, this paper jointly makes two factor matrices nonparametric and sparse, which could be applied to broader scenarios, such as co-clustering. This paper is seen to be much more flexible than Gaussian process-based and hierarchial Beta process-based dIBPs in terms of allowing the two corresponding binary matrix columns to have greater variations in their nonzero entries. Our experiments on synthetic data show the merits of this paper compared with the state-of-the-art models in respect of factorization efficiency, sparsity, and flexibility. Experiments on real-world data sets demonstrate the efficiency of this paper in document-word co-clustering tasks.
© 2013 IEEE. The explosive increasing of the social media data on the Web has created and promoted the development of the social media big data mining area welcomed by researchers from both academia and industry. The sentiment computing of news event is a significant component of the social media big data. It has also attracted a lot of researches, which could support many real-world applications, such as public opinion monitoring for governments and news recommendation for Websites. However, existing sentiment computing methods are mainly based on the standard emotion thesaurus or supervised methods, which are not scalable to the social media big data. Therefore, we propose an innovative method to do the sentiment computing for news events. More specially, based on the social media data (i.e., words and emoticons) of a news event, a word emotion association network (WEAN) is built to jointly express its semantic and emotion, which lays the foundation for the news event sentiment computation. Based on WEAN, a word emotion computation algorithm is proposed to obtain the initial words emotion, which are further refined through the standard emotion thesaurus. With the words emotion in hand, we can compute every sentence's sentiment. Experimental results on real-world data sets demonstrate the excellent performance of the proposed method on the emotion computing for news events.
Liu, W, Luo, X, Xuan, J, Jiang, D & Xu, Z 2017, 'ASSOCIATION LINK NETWORK BASED SEMANTIC COHERENCE MEASUREMENT FOR SHORT TEXTS OF WEB EVENTS', JOURNAL OF WEB ENGINEERING, vol. 16, no. 1-2, pp. 39-62.
Ma, W, Luo, X, Xuan, J, Xue, R & Guo, Y 2017, 'Discover semantic topics in patents within a specific domain', Journal of Web Engineering, vol. 16, no. 7-8, pp. 653-675.
© Rinton Press. Patent topic discovery is critical for innovation-oriented enterprises to hedge the patent application risks and raise the success rate of patent application. Topic models are commonly recognized as an efficient tool for this task by researchers from both academy and industry. However, many existing well-known topic models, e.g., Latent Dirichlet Allocation (LDA), which are particularly designed for the documents represented by word-vectors, exhibit low accuracy and poor interpretability on patent topic discovery task. The reason is that 1) the semantics of documents are still under-explored in a specific domain 2) and the domain background knowledge is not successfully utilized to guide the process of topic discovery. In order to improve the accuracy and the interpretability, we propose a new patent representation and organization with additional inter-word relationships mined from title, abstract, and claim of patents. The representation can endow each patent with more semantics than word-vector. Meanwhile, we build a Backbone Association Link Network (Backbone ALN) to incorporate domain background semantics to further enhance the semantics of patents. With new semantic-rich patent representations, we propose a Semantic LDA model to discover semantic topics from patents within a specific domain. It can discover semantic topics with association relations between words rather than a single word vector. At last, accuracy and interpretability of the proposed model are verified on real-world patents datasets from the United States Patent and Trademark Office. The experimental results show that Semantic LDA model yields better performance than other conventional models (e.g., LDA). Furthermore, our proposed model can be easily generalized to other related text mining corpus.
Xu, Z, Liu, Y, Xuan, J, Chen, H & Mei, L 2017, 'Crowdsourcing based social media data analysis of urban emergency events', MULTIMEDIA TOOLS AND APPLICATIONS, vol. 76, no. 9, pp. 11567-11584.View/Download from: Publisher's site
Xu, Z, Xuan, J, Liu, Y, Choo, K-KR, Mei, L & Hu, C 2017, 'Building spatial temporal relation graph of concepts pair using web repository', INFORMATION SYSTEMS FRONTIERS, vol. 19, no. 5, pp. 1029-1038.View/Download from: Publisher's site
Xu, Z, Zhang, H, Hu, C, Liu, Y, Xuan, J & Mei, L 2017, 'Crowdsourcing-based timeline description of urban emergency events using social media', INTERNATIONAL JOURNAL OF AD HOC AND UBIQUITOUS COMPUTING, vol. 25, no. 1-2, pp. 41-51.View/Download from: Publisher's site
Xu, Z, Zhang, H, Hu, C, Liu, Y, Xuan, J & Mei, L 2017, 'Crowdsourcing-based timeline description of urban emergency events using social media', International Journal of Ad Hoc and Ubiquitous Computing, vol. 25, no. 1-2, pp. 41-51.View/Download from: Publisher's site
Copyright © 2017 Inderscience Enterprises Ltd. Crowdsourcing is a newly emerging service platform and business model in the Internet. Analysis and description about urban emergency events, e.g., fires, storms and traffic jams are of great importance to protect the security of humans. Recently, social media feeds are rapidly emerging as a novel platform for providing and dissemination of information that is often geographic. In this paper, in order to describe the timeline of real-time urban emergency events, the new web mining task timeline description (TD) is proposed. Firstly, the related information of an urban emergency event is extracted from Weibo messages. Secondly, the valid message including the semantic or spatial information is detected in this step. Thirdly, detected valid messages are used to build the TD. Case studies on real datasets show the proposed model has good performance and high effectiveness in the analysis and description of urban emergency events.
Xuan, J, Luo, X, Lu, J & Zhang, G 2017, 'Explicitly and implicitly exploiting the hierarchical structure for mining website interests on news events', INFORMATION SCIENCES, vol. 420, pp. 263-277.View/Download from: Publisher's site
Lu, J, Xuan, J, Zhang, G, Xu, YD & Luo, X 2017, 'Bayesian Nonparametric Relational Topic Model through Dependent Gamma Processes', IEEE Transactions on Knowledge and Data Engineering, vol. 29, no. 7, pp. 1357-1369.View/Download from: Publisher's site
Traditional relational topic models provide a successful way to discover the hidden topics from a document network. Many
theoretical and practical tasks, such as dimensional reduction, document clustering, and link prediction, could benefit from this revealed
knowledge. However, existing relational topic models are based on an assumption that the number of hidden topics is known a priori,
which is impractical in many real-world applications. Therefore, in order to relax this assumption, we propose a nonparametric relational
topic model using stochastic processes instead of fixed-dimensional probability distributions in this paper. Specifically, each document
is assigned a Gamma process, which represents the topic interest of this document. Although this method provides an elegant solution,
it brings additional challenges when mathematically modeling the inherent network structure of typical document network, i.e., two
spatially closer documents tend to have more similar topics. Furthermore, we require that the topics are shared by all the documents. In
order to resolve these challenges, we use a subsampling strategy to assign each document a different Gamma process from the global
Gamma process, and the subsampling probabilities of documents are assigned with a Markov Random Field constraint that inherits the
document network structure. Through the designed posterior inference algorithm, we can discover the hidden topics and its number
simultaneously. Experimental results on both synthetic and real-world network datasets demonstrate the capabilities of learning the
hidden topics and, more importantly, the number of topics.
Liu, W, Luo, X, Gong, Z, Xuan, J, Kou, NM & Xu, Z 2016, 'Discovering the core semantics of event from social media', Future Generation Computer Systems, vol. 64, pp. 175-185.View/Download from: Publisher's site
Xu, Z, Zhang, H, Hu, C, Mei, L, Xuan, J, Choo, K-KR, Sugumaran, V & Zhu, Y 2016, 'Building knowledge base of urban emergency events based on crowdsourcing of social media', Concurrency and Computation: Practice and Experience, vol. 28, no. 15, pp. 4038-4052.View/Download from: Publisher's site
Xuan, J, Luo, X, Zhang, G, Lu, J & Xu, Z 2016, 'Uncertainty Analysis for the Keyword System of Web Events', IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, vol. 46, no. 6, pp. 829-842.View/Download from: Publisher's site
Luo, X, Xuan, J, Lu, J & Zhang, G 2016, 'Measuring the Semantic Uncertainty of News Events for Evolution Potential Estimation', ACM TRANSACTIONS ON INFORMATION SYSTEMS, vol. 34, no. 4.View/Download from: Publisher's site
Liu, Y, Luo, X & Xuan, J 2015, 'Online hot event discovery based on Association Link Network', Concurrency and Computation: Practice and Experience, vol. 27, no. 15, pp. 4001-4014.View/Download from: Publisher's site
Ma, Q, Luo, X, Xuan, J & Liu, H 2015, 'BAYESIAN-BASED TYPE DISCRIMINATION OF WEB EVENTS', JOURNAL OF WEB ENGINEERING, vol. 14, no. 5-6, pp. 525-544.
Incorporating the side information of text corpus, i.e., authors,
time stamps, and emotional tags, into the traditional
text mining models has gained significant interests in the
area of information retrieval, statistical natural language
processing, and machine learning. One branch of these works
is the so-called Author Topic Model (ATM), which incorporates
the authors's interests as side information into the
classical topic model. However, the existing ATM needs to
predefine the number of topics, which is difficult and inappropriate
in many real-world settings. In this paper, we propose
an Infinite Author Topic (IAT) model to resolve this
issue. Instead of assigning a discrete probability on fixed
number of topics, we use a stochastic process to determine
the number of topics from the data itself. To be specific, we
extend a gamma-negative binomial process to three levels in
order to capture the author-document-keyword hierarchical
structure. Furthermore, each document is assigned a mixed
gamma process that accounts for the multi-author's contribution
towards this document. An efficient Gibbs sampling
inference algorithm with each conditional distribution being
closed-form is developed for the IAT model. Experiments
on several real-world datasets show the capabilities of our
IAT model to learn the hidden topics, authors' interests on
these topics and the number of topics simultaneously.
Graph mining has been a popular research area
because of its numerous application scenarios. Many unstructured
and structured data can be represented as graphs, such as,
documents, chemical molecular structures, and images. However,
an issue in relation to current research on graphs is that they
cannot adequately discover the topics hidden in graph-structured
data which can be beneficial for both the unsupervised learning
and supervised learning of the graphs. Although topic models
have proved to be very successful in discovering latent topics,
the standard topic models cannot be directly applied to graphstructured
data due to the "bag-of-word" assumption. In this
paper, an innovative graph topic model (GTM) is proposed to
address this issue, which uses Bernoulli distributions to model
the edges between nodes in a graph. It can, therefore, make the
edges in a graph contribute to latent topic discovery and further
improve the accuracy of the supervised and unsupervised learning
of graphs. The experimental results on two different types
of graph datasets show that the proposed GTM outperforms the
latent Dirichlet allocation on classification by using the unveiled
topics of these two models to represent graphs.
Luo, X, Xuan, J & Liu, H 2014, 'WEB EVENT STATE PREDICTION MODEL: COMBINING PRIOR KNOWLEDGE WITH REAL TIME DATA', JOURNAL OF WEB ENGINEERING, vol. 13, no. 5-6, pp. 483-506.
Lin, A, Lu, J, Xuan, J, Zhu, F & Zhang, G 2019, 'One-stage deep instrumental variable method for causal inference from observational data', Proceedings - IEEE International Conference on Data Mining, ICDM, IEEE International Conference on Data Mining, IEEE, Beijing, China, pp. 419-428.View/Download from: Publisher's site
© 2019 IEEE. Causal inference from observational data aims to estimate causal effects when controlled experimentation is not feasible, but it faces challenges when unobserved confounders exist. The instrumental variable method resolves this problem by introducing a variable that is correlated with the treatment and affects the outcome only through the treatment. However, existing instrumental variable methods require two stages to separately estimate the conditional treatment distribution and the outcome generating function, which is not sufficiently effective. This paper presents a one-stage approach to jointly estimate the treatment distribution and the outcome generating function through a cleverly designed deep neural network structure. This study is the first to merge the two stages to leverage the outcome to the treatment distribution estimation. Further, the new deep neural network architecture is designed with two strategies (i.e., shared and separate) of learning a confounder representation account for different observational data. Such network architecture can unveil complex relationships between confounders, treatments, and outcomes. Experimental results show that our proposed method outperforms the state-of-the-art methods. It has a wide range of applications, from medical treatment design to policy making, population regulation and beyond.
Xu, Z, Xuan, J, Zhu, Y & Wei, X 2016, 'Building the profile of web events based on website measurement', FC 2016: Frontier Computing (Lecture Notes in Electrical Engineering), International Conference on Frontier Computing, Springer, FC, pp. 3-10.View/Download from: Publisher's site
© Springer Nature Singapore Pte Ltd. 2018. Nowadays, Web makes it possible to study emergencies from web information due to its real-time, open, and dynamic features. After the emergence of a web event, there will be numerous websites publishing webpages to cover this web event. Measuring temporal features in evolution course of web events can help people timely know and understand which events are emergencies, so harms to the society caused by emergencies can be reduced. In this paper, website preference is formally defined and mined by three proposed strategies which are all explicitly or implicitly based on the three-level networks: website-level, webpage-level and keyword-level. An iterative algorithm is firstly introduced to calculate outbreak power of web events, and increased web pages of events, increased attributes of events, distribution of attributes in web pages and the relationships of attributes are embedded into this iterative algorithm as the variables. By means of prior knowledge, membership grade of web events belong to each type can be calculated, and then the type of web events can be discriminated. Experiments on real data set demonstrate the proposed algorithm is both efficient and effective, and it is capable of providing accurate results of discrimination.
Lin, A, Xuan, J, Zhang, G & Lu, J 2018, 'Causal inference with Gaussian processes for support of terminating or maintaining an existing program', Data Science and Knowledge Engineering for Sensing Decision Support, Conference on Data Science and Knowledge Engineering for Sensing Decision Support (FLINS 2018), WORLD SCIENTIFIC, Belfast, Northern Ireland.View/Download from: Publisher's site
Zhang, Y, Wang, W, Xuan, J, Lu, J, Zhang, G & Lin, H 2018, 'Map-based medical practice behavior analysis: Methodology and a case study on Australia's medical practices', the 13th International FLINS Conference, pp. 1323-1330.
Liu, Q, Huang, H, Zhang, G, Gao, Y, Xuan, J & Lu, J 2018, 'Semantic structure-based word embedding by incorporating concept convergence and word divergence', 32nd AAAI Conference on Artificial Intelligence, AAAI 2018, pp. 5261-5268.
Copyright © 2018, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Representing the semantics of words is a fundamental task in text processing. Several research studies have shown that text and knowledge bases (KBs) are complementary sources for word embedding learning. Most existing methods only consider relationships within word-pairs in the usage of KBs. We argue that the structural information of well-organized words within the KBs is able to convey more effective and stable knowledge in capturing semantics of words. In this paper, we propose a semantic structure-based word embedding method, and introduce concept convergence and word divergence to reveal semantic structures in the word embedding learning process. To assess the effectiveness of our method, we use WordNet for training and conduct extensive experiments on word similarity, word analogy, text classification and query expansion. The experimental results show that our method outperforms state-of-the-art methods, including the methods trained solely on the corpus, and others trained on the corpus and the KBs.
Xu, Z & Xuan, J 2015, 'Temporal Learning of Semantic Relations between Concepts Using Web Repository.', Proceedings of the 2015 11th International Conference on Semantics, Knowledge and Grids (SKG), International Conference on Semantics, Knowledge and Grids, IEEE Computer Society, Beijing, China, pp. 239-243.View/Download from: Publisher's site
In this paper, the study on generating temporal learning of relations between concepts is proposed. The purpose of the proposed study is to annotate a relation between concepts with semantic, temporal, concise, and structured information, which can release the cognitive burden of learning relations between concepts for users. The temporal annotations can help users to learn and understand the unfamiliar or new emerged relations between concepts. A general method is proposed to generate temporal learning of a relation between concepts by constructing its relation words, relation sentences, relation graph, and relation factor. Empirical experiments on movie star dataset show that the proposed algorithm is effective and accurate
Xuan, J, Lu, J, Zhang, G, Xu, RYD & Luo, X 2015, 'Infinite author topic model based on mixed gamma-negative binomial process', Proceedings - IEEE International Conference on Data Mining, ICDM, IEEE International Conference on Data Mining, IEEE, Atlantic City, USA, pp. 489-498.View/Download from: Publisher's site
Incorporating the side information of text corpus, i.e., authors, time stamps, and emotional tags, into the traditionaltext mining models has gained significant interests in the area of information retrieval, statistical natural language processing, andmachine learning. One branch of these works is the so-called Author Topic Model (ATM), which incorporates the authors'sinterests as side information into the classical topic model. However, the existing ATM needs to predefine the number of topics, which is difficult and inappropriate in many real-world settings. In this paper, we propose an Infinite Author Topic (IAT) modelto resolve this issue. Instead of assigning a discrete probability on fixed number of topics, we use a stochastic process to determinethe number of topics from the data itself. To be specific, we extend a gamma-negative binomial process to three levels in orderto capture the author-document-keyword hierarchical structure. Furthermore, each document is assigned a mixed gamma processthat accounts for the multi-author's contribution towards this document. An efficient Gibbs sampling inference algorithm witheach conditional distribution being closed-form is developed for the IAT model. Experiments on several real-world datasets showthe capabilities of our IAT model to learn the hidden topics, authors' interests on these topics and the number of topicssimultaneously.
Jing, J, Luo, X, Xuan, J & Liu, W 2014, 'Cognition-Based Semantic Annotation for Web Images.', BDCloud, International Conference on Big Data and Cloud Computing, IEEE, Sydney, NSW, Australia, pp. 540-546.View/Download from: Publisher's site
Due to the semantic gap between low-level visual features and high-level semantic content of images, the methods for image annotation based on low-level visual features, cannot well meet the requirement of knowledge discovery from web images. Therefore, the automatic acquisition for high-level semantic content of image has become a hot research topic. The traditional image annotation methods represent images only by a few keywords, which cannot completely describe and rationally organize the high-level semantics of images, so it will lose a great deal of semantic information. Based on the different levels and different aspects of web images, we propose a new method to express and organize the high-level semantic content of web images. The method expresses the different levels semantic content of one image as a three-level network, composed of background semantic level, complementary semantic level and fine-grained semantic level. The experimental results show that our method is effective and efficient on the image annotation.
Junyu Xuan, Jie Lu, Guangquan Zhang & Xiangfeng Luo 2014, 'Extension of similarity measures in VSM: From orthogonal coordinate system to affine coordinate system', Neural Networks (IJCNN), 2014 International Joint Conference on, IEEE International Joint Conference on Neural Networks, IEEE, Beijing, China, pp. 4084-4091.View/Download from: Publisher's site
Similarity measures are the foundations of many research areas, e.g. information retrieval, recommender system and machine learning algorithms. Promoted by these application scenarios, a number of similarity measures have been proposed and proposing. In these state-of-the-art measures, vector-based representation is widely accepted based on Vector Space Model (VSM) in which an object is represented as a vector composed of its features. Then, the similarity between two objects is evaluated by the operations on two corresponding vectors, like cosine, extended jaccard, extended dice and so on. However, there is an assumption that the features are independent of each others. This assumption is apparently unrealistic, and normally, there are relations between features, i.e. the co-occurrence relations between keywords in text mining area. In this paper, a space geometry-based method is proposed to extend the VSM from the orthogonal coordinate system (OVSM) to affine coordinate system (AVSM) and OVSM is proved to be a special case of AVSM. Unit coordinate vectors of AVSM are inferred by the relations between features which are considered as angles between these unit coordinate vectors. At last, five different similarity measures are extended from OVSM to AVSM using unit coordinate vectors of AVSM. Within the numerous application fields of similarity measures, the task of text clustering is selected to be the evaluation criterion. Documents are represented as vectors in OVSM and AVSM, respectively. The clustering results show that AVSM outweighs the OVSM.
Xuan, J, Lu, J, Zhang, G & Luo, X 2014, 'Release 'Bag-of-Words' Assumption of Latent Dirichlet Allocation', FOUNDATIONS OF INTELLIGENT SYSTEMS (ISKE 2013), International Conference on Intelligent Systems and Knowledge Engineering, Springer Berlin Heidelberg, Shenzhen, PEOPLES R CHINA, pp. 83-92.View/Download from: Publisher's site
Xuan, J, Luo, X & Lu, J 2013, 'Mining Websites Preferences on Web Events in Big Data Environment', 2013 IEEE 16th International Conference on Computational Science and Engineering (CSE), International Conference on Computational Science and Engineering, IEEE, Sydney, Australia, pp. 1043-1050.View/Download from: Publisher's site
On the web, there are numerous websites publishing web pages to cover the events occurring in society. The web events data satisfies the well-accepted attributes of big data: Volume, Velocity, Variety and Value. As a great value of web events data, website preferences can help the followers of web events, e.g. peoples or organizations, to select the proper websites to follow their interested aspects of web events. However, the big volume, fast evolution speed, multisource and unstructured data all together make the value of website preferences mining very challenging. In this paper, website preference is formally defined at first. Then, according to the hierarchical attribute of web events data, we propose a hierarchical network model to organize big data of a web event from different organizations, different areas and different nations at a given time stamp. With this hierarchical network structure in hand, two strategies are proposed to mine the value of websites preferences from web events data. The first straightforward strategy utilizes the communities of keyword level network and the mapping relations between websites and keywords to unveil the Value in them. By taking the whole hierarchical network structure into consideration, an iterative algorithm is proposed in second strategy to refine the keyword communities like the first strategy. At last, an evaluation criteria of website preferences is designed to compare the performances of two proposed strategies. Experimental results show the proper combination of horizontal relations (each level network) with vertical relations (mapping relations between three level networks) can extract more value from web events data and then improve the efficiency on website preferences mining.
Xuan, J, Luo, X, Zhang, S, Xu, Z, Liu, H & Ye, F 2011, 'Building Hierarchical Keyword Level Association Link Networks for Web Events Semantic Analysis.', 2011 Ninth IEEE International Conference on Dependable, Autonomic and Secure Computing, International Conference on Dependable, Autonomic and Secure Computing, IEEE Computer Society, Sydney, NSW, Australia, pp. 987-994.View/Download from: Publisher's site
With the increase of information scale of web events on the time, it is extremely difficult and challenging to grasp the semantics of web events artificially, because of the limitation of the time and energy of human beings. Herein, we propose a method to map the web event to keyword level association link network (KALN) for deep analysis of the semantics of web events, such as the evolution semantics of web events. Firstly, the original KALN is constructed at a given time by traditional data mining technologies. Then, the hierarchical KALN, consisted of Theme Layer Network, Backbone Layer Network and Tidbit Layer Network, is built based on the original KALN by information entropy to identify the different semantic levels of the web event, including stable semantics, sub-stable semantics and unstable semantics. With the semantic analysis of hierarchical KALN, human could easily gain a thorough understanding of the web event. Finally, experiments show that our method can effectively capture the different level semantics of web events.
Zhang, S, Luo, X, Xuan, J, Chen, X & Xu, W 2011, 'Discovering small-world in Association Link Networks for Web-based learning', MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops - ACM International Workshop on Multimedia Technologies for Distance Learning, MTDL'11, pp. 19-24.View/Download from: Publisher's site
Association Link Network (ALN) is a kind of Semantic Link Network built by mining the association relations among Web resources for effectively supporting Web intelligent application such as Web-based learning, and knowledge acquisition. This paper explores the Small-World properties of ALN to provide theoretical support for Web-based learning. First, a filtering algorithm of ALN is proposed to generate the filtered status of ALN by adjusting the filtering parameter. Secondly, the Small-World properties of ALN at the filtered status are calculated and analyzed by regression analysis to observe the changing trend of Small-World properties. After that, comparison of the Small-World properties between ALN and random graph shows that ALN reveals prominent Small-World characteristic. The discovery of Small-World characteristic of ALN can provide theoretical support for Web-based learning. © 2011 ACM.
Xuan, J, Lu, J, Zhang, G, Xu, RYD & Luo, X 2015, 'Dependent Indian Buffet Process-based Sparse Nonparametric Nonnegative Matrix Factorization'.
Nonnegative Matrix Factorization (NMF) aims to factorize a matrix into two
optimized nonnegative matrices appropriate for the intended applications. The
method has been widely used for unsupervised learning tasks, including
recommender systems (rating matrix of users by items) and document clustering
(weighting matrix of papers by keywords). However, traditional NMF methods
typically assume the number of latent factors (i.e., dimensionality of the
loading matrices) to be fixed. This assumption makes them inflexible for many
applications. In this paper, we propose a nonparametric NMF framework to
mitigate this issue by using dependent Indian Buffet Processes (dIBP). In a
nutshell, we apply a correlation function for the generation of two stick
weights associated with each pair of columns of loading matrices, while still
maintaining their respective marginal distribution specified by IBP. As a
consequence, the generation of two loading matrices will be column-wise
(indirectly) correlated. Under this same framework, two classes of correlation
function are proposed (1) using Bivariate beta distribution and (2) using
Copula function. Both methods allow us to adopt our work for various
applications by flexibly choosing an appropriate parameter settings. Compared
with the other state-of-the art approaches in this area, such as using Gaussian
Process (GP)-based dIBP, our work is seen to be much more flexible in terms of
allowing the two corresponding binary matrix columns to have greater variations
in their non-zero entries. Our experiments on the real-world and synthetic
datasets show that three proposed models perform well on the document
clustering task comparing standard NMF without predefining the dimension for
the factor matrices, and the Bivariate beta distribution-based and Copula-based
models have better flexibility than the GP-based model.