Title: Optimization of modular granular neural networks using bio-inspired optimization algorithms with fuzzy parameter adaptation for human recognition

Spearker: Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexico

Abstract: In this talk, new models of modular neural networks optimized with bio-inspired optimization algorithms using fuzzy parameter adaptation are proposed. The models use a granular approach based on analyzing database complexity. In this case the proposed method is tested with the problem of human recognition based on face information. The ORL and the ESSEX face databases are used to test the effectiveness of the proposed method. To compare with other related works using the same databases, four cases are established (3 for the ESSEX Database and 1 for the ORL Database). The results using the proposed method are better than the results achieved by other works, and this affirmation is based on a statistical comparison of results. The main idea is to design the architectures of modular neural networks using bio-inspired optimization methods, such as particle swarm optimization, grey wolf optimization and the firefly algorithm. The distribution of persons in each granule is determined by an initial analysis, resulting in a grouping of data with the same complexity. The proposed method allows the optimization of multiple modular neural networks that use different sizes of data sets for the training phase, which means that multiple results can be obtained.

Biography: Prof. Patricia Melin holds the Doctor in Science degree (Doctor Habilitatus D.Sc.) in Computer Science from the Polish Academy of Sciences (with the Dissertation “Hybrid Intelligent Systems for Pattern Recognition using Soft Computing”). She is a Professor of Computer Science in the Graduate Division, Tijuana Institute of Technology, Tijuana, Mexico, since 1998. In addition, she is serving as Director of Graduate Studies in Computer Science and head of the research group on Hybrid Neural Intelligent Systems (2000-present). Currently, she is President of NAFIPS (North American Fuzzy Information Processing Society). Prof. Melin is the founding Chair of the Mexican Chapter of the IEEE Computational Intelligence Society. She is member of the IEEE Neural Network Technical Committee (2007 to present), the IEEE Fuzzy System Technical Committee (2014 to present) and in Chair of the Task Force on Hybrid Intelligent Systems (2007 to present) and she is currently Associate Editor of the Journal of Information Sciences and IEEE Transactions on Fuzzy Systems. She is member of NAFIPS, IFSA, and IEEE. She belongs to the Mexican Research System with level III. Her research interests are in Modular Neural Networks, Type-2 Fuzzy Logic, Pattern Recognition, Fuzzy Control, Neuro-Fuzzy and Genetic-Fuzzy hybrid approaches. She has published over 220 journal papers, 10 authored books, 20 edited books, and more than 250 papers in conference proceedings with h-index of 42. She has served as Guest Editor of several Special Issues in the past, in journals like: Applied Soft Computing, Intelligent Systems, Information Sciences, Non-Linear Studies, JAMRIS, Fuzzy Sets and Systems, and Engineering Letters.

Title: Opinion Mining and Sentiment Analysis with Deep Learning Techniques

Spearker: Alexander Gelbukh, Instituto Politécnico Nacional, Mexico City, Mexico

Abstract:Opinion mining is an active area of research that has recently attracted much interest from research community, industry, government, and political circles. Opinion mining aims to process the huge body of user-contributed content in Internet and social networks to extract user’s opinions, suggestions, and feedback on a very wide range of products, services, issues, or political figures and use this information to help decision making in industry and politics, as well as to help consumers to make informed purchasing decisions. Sentiment analysis is the main subtask of opinion mining. It consists in determining whether a given opinion (such as review, tweet, or a video clip) is positive or negative or what emotion it expresses. Other subtasks of opinion mining include author profiling and deception detection. I will present some deep-learning architectures developed in our group for utterance-level sentiment analysis in texts and video clips, as well as for determining the psychological type of the author of a text.

Biography: Prof. Alexander Gelbukh holds MSc (mathematics) degree by the Moscow State “Lomonosov” University, Russia, and PhD (computer science) degree by the All-Russian Institute for Scientific and Technical Information, Russia. He is Research Professor and Head of the Natural Language Processing Laboratory of the Center for Computing Research of the Instituto Politécnico Nacional, Mexico, Honorary Professor of the Amity University, India, and Invited professor of the National University of Colombia. He has been Invited researcher of Waseda University, Japan, and Distinguished Visiting Professor of Chung-Ang University, Korea. He is a member of the Mexican Academy of Sciences and founding member of the Mexican Academy of Computing. He has been the President of the Mexican Society for Artificial Intelligence and of the Mexican Association of Natural Language Processing. His main areas of interest include computational linguistics and artificial intelligence. He is an author or co-author of more than 500 research publications, including 8 books, editor of more than 80 books, proceeding volumes, or journal special issues, and editor-in-chief or member of editorial board for more than 10 international journals. He is the founder and chair of the CICLing series of international conferences. He was a Chair, Honorary Chair, or Program Committee Chair of more than 50 international conferences. He has been advisor more than 25 PhD students. See details at www.Gelbukh.com.