Marie-Jeanne Lesot
Associate Professor
Department of Computer Science
Lab of Paris 6 (LIP6)
Data Explainability Through Linguistic Expression of Extracted Knowledge
The pervasive use of data science techniques extracts regularities from available data for different tasks, such as prediction, characterisation or structuring. A current challenge is to improve the legibility of the obtained results, so as to allow a data expert to understand better the content of the data. One way to address this challenge consists in presenting them in natural language, offering linguistic expressions which may be easier to interpret for the user. The choice of such a result formulation then has an impact on the machine learning techniques to be applied to the data.
The talk will illustrate these questions for numerical data as well as for time series, respectively discussing the extraction of gradual itemsets, that linguistically express knowledge about feature covariations, and the extraction of periodicity-related linguistic summaries, using the specific quantifier “regularly”. In both cases, as well as for enriched contextual variants, the question is to define precisely the associated semantics and to design efficient extraction algorithms. The talk will also discuss the issue of measuring the relevance of the linguistic terms used to express the summaries, both with respect to the data structure, in case of linguistic variables, and with respect to the cognitive interpretation, in case of approximate numerical expressions.
Biography
Marie-Jeanne Lesot obtained her PhD in 2005 and her habilitation in 2013 from the University Pierre et Marie Curie (Paris, France). In 2005, she was a post-doctoral fellow at Otto-von-Guericke-University (Magdeburg, Germany) during one year. Since 2006, she has been an associate professor in the department of Computer Science Lab of Paris 6 (LIP6) and a member of the Learning and Fuzzy Intelligent systems (LFI) group. Her research interests focus on fuzzy machine learning with an objective of data interpretation and semantics integration and, in particular, to model and manage subjective information; they include similarity measures, fuzzy clustering, linguistic summaries, affective computing and information scoring.
Publications: http://webia.lip6.fr/~lesot/publications.html
Lawrence O. Hall
Department of Computer Science and Engineering
University of South Florida
Explorations in BIG Data and sMall Data with a Fuzzy Perspective
Consider looking for interactions in and between social networks such as Github, Twitter and Reddit. If, at the hourly level, you want to predict events or cascades or the effect of one entity on another, there is an enormous amount of data to analyze. Now, consider non-invasive medical image data, there is never as much data easily available as one would prefer. However, the data can be used to make useful predictions about prognosis and treatments. This talk will discuss approaches using (deep) learning to do predictions from big social network data and smaller medical image data sets, with successes, pitfalls and how fuzzy approaches can help. Some examples are for big data, even when labels exist, there are likely groups to be discovered by scalable fuzzy clustering. Temporal message and event boundaries are fuzzy with sent and “actually” sent times having some spread due to system delays. This talk will discuss challenges, results, and opportunities to improve performance using fuzzy techniques.
Biography
LAWRENCE O. HALL is a Distinguished University Professor in the Department of Computer Science and Engineering at University of South Florida. He received his Ph.D. in Computer Science from the Florida State University in 1986 and a B.S. in Applied Mathematics from the Florida Institute of Technology in 1980. He is a fellow of the IEEE. He is a fellow of the AAAS, AIMBE and IAPR. He received the Norbert Wiener award in 2012 and the Joseph Wohl award in 2017 from the IEEE SMC Society. His research interests lie in learning from big data, distributed machine learning, medical image understanding, bioinformatics, pattern recognition, modeling impression in decision making, and integrating AI into image processing. He continues to explore un and semi-supervised learning using scalable fuzzy approaches. He has authored or co-authored over 90 publications in journals, as well as many conference papers and book chapters. He has received over $5M in research funding from agencies such as the National Science Foundation, National Institutes of Health, Department of Energy, DARPA, and NASA.
Hisao Ishibuchi
Department of Computer Science and Engineering
Southern University of Science and Technology (SUSTech), China
Fuzzy Rule-Based Classifier Design: Accuracy, Interpretability and Explanation Ability
Fuzzy rule-based system design involves conflicting objectives such as interpretability maximization and accuracy maximization. For example, linguistic interpretability of fuzzy rule-based systems can be improved by decreasing the number of fuzzy rules, the number of antecedent conditions in each fuzzy rule, and the complexity of fuzzy partition for each input variable. However, those interpretability improvement efforts often degrade the accuracy of fuzzy rule-based systems. That is, very simple fuzzy rule-based systems with high linguistic interpretability usually do not have high accuracy. In the 1990s, conflicting objectives were combined into a single integrated objective function, which was optimized by a single-objective optimization algorithm. Currently those objectives are handled as different objectives and simultaneously optimized by an evolutionary multi-objective optimization (EMO) algorithm. A large number of non-dominated fuzzy rule-based systems can be obtained by a single run of an EMO algorithm. In this talk, we focus on fuzzy rule-based classifier design. This talk starts with a brief review of well-known highly-cited studies related to fuzzy rule-based classifier design from two viewpoints: Accuracy maximization and interpretability maximization. Next, fuzzy rule-based classifier design is explained as multi-objective optimization problems to which EMO algorithms are applied. Then, we discuss the explanation ability of fuzzy rule-based classifiers, which is the ability to explain in a human understandable manner why an input pattern is classified as an output class. Finally, we discuss future research directions in fuzzy rule-based classifier design including deep fuzzy classifiers.
Biography
Hisao Ishibuchi is currently a Chair Professor at the Department of Computer Science and Engineering, Southern University of Science and Technology (SUSTech), Shenzhen, China. After completing his MS studies on precision mechanics in Kyoto University in March 1987, he joined Prof. Hideo Tanaka’s research group as a Research Assistant, where he started his research on fuzzy systems. Since then, he had been with Osaka Prefecture University for 30 years. In April 2017, he moved to China to join SUSTech.
In the early 1990s, he proposed a heuristic fuzzy rule generation method for pattern classification problems. Then he combined a genetic algorithm with the rule generation method to select a small number of fuzzy rules with high classification ability. He generalized this approach to multi-objective genetic rule selection in the 1990s, which was the first study on multi-objective genetic fuzzy systems. In the 2000s, he proposed multi-objective fuzzy genetics-based machine learning. He received Best Paper Awards from FUZZ-IEEE 2009, 2011 and GECCO 2004, 2017, 2018. He also received a JSPS Prize from Japan Society for the Promotion of Science in 2007 and a Fuzzy Systems Pioneer Award from the IEEE CIS in 2019. In 2018, he was selected in the “Recruitment Program of Global Experts for Foreign Experts” known as the “Thousand Talents Program” in China.