Keynote Speakers

Prof. Plamen Angelov

Title: Empirical Fuzzy Sets and Systems – a paradigm shift

Abstract: We witness an exponential growth in the scale and complexity of the data sets and streams being generated by sensors, people, society, industry, etc. This is being increasingly seen as an untapped resource which offers new opportunities for extracting aggregated information to inform decision-making in policy and commerce. The traditional Fuzzy Sets Theory and the Fuzzy Systems have been defined over 50 years ago in the seminal paper by Professor Lotfi Zadeh [1] and now matured. In 1990s in addition to the traditional subjective way of designing fuzzy sets the so-called data driven design method started to be popular and was developed. The subjective approach has its own very strong rationale in the two way process of: i) extracting expert knowledge and representing it in a mathematical form through the membership functions, and ii) the ability to represent and extract form data human intelligible and understandable, transparent linguistic information in the form of IF …THEN rules. In addition, since mid-1970s (Mamdani or Zadeh-Mamdani) and since mid-1980s (Takagi-Sugeno) fuzzy rule-based (FRB) systems started to be developed and are now widely applied. Although, there are other types of fuzzy systems (relational, etc.) one particular type of FRB systems which was introduced recently by Angelov and Yager [2] called AnYa deserves a special attention. Both Mamdani and Takagi-Sugeno type of FRB share the exact same antecedent part (the IF) and only (although significantly) differ by the consequent (THEN) part. AnYa type FRB, however, has a quite different antecedent (IF) part. The main issue in the design of the fuzzy sets and systems is the very fundamental one – the membership function by which they are defined in first place. It is practically very difficult and controversial to define membership functions both form experts and from data. This is also related to the more general issue of assumptions made and handcrafting which machine learning (including statistical methods) are facing and is now hotly researched.

In this talk a new approach [3] will be discussed which leads to a new kind of fuzzy sets and systems – empirical fuzzy sets and FRB systems (εFS and εFRB). It is grounded at the recently introduced more general concept and a computational framework of Empirical Data Analytics (EDA) [4]. In this talk, EDA will be introduced and then on its basis the εFS and εFRB will be defined. It will be presented how εFS and εFRB allow preserving the subjective specifics that fuzzy sets and systems are strong with. At the same time, it will be shown how εFS and εFRB can benefit from the vast amount of data that may be available. For example, εFS and εFRB will still allow extracting expert knowledge by questionnaires or other forms, but will make this much more easy for the expert and not ambiguous (the experts will not be asked to define membership values or parameters, but only (optionally) the labels/names of the linguistic terms, classes (if any). For example, if we chose a car, we can simply say which one we like (or possibly how much), but we do not need to specify why or define per feature (price, max speed, etc.). Moreover, with these new type of εFS and εFRB one can tackle heterogeneous data and combine categorical (e.g. gender, occupation, number of doors) with continuous variables like price, max speed, size, etc.

In the talk, it will be demonstrated how on the basis of εFS and εFRB one can build empirically fuzzy classifiers (εF Classifiers), predictors (εF Predictors), controllers (εF Controllers), recommender systems, etc. Moreover, these can be evolving, not just fixed structure. This will allow studying the dynamic changes in human preferences as well as to build more efficient recommender systems where the only necessary input form the users is the preference (“likes” or “retweets” or “clicks”).

References:

[1] L. A. Zadeh, Fuzzy Sets, Information and Control, 8(3): 338-353, 1965.
[2] P. Angelov and R. Yager, A new type of simplified fuzzy rule-based system, International Journal on General Systems, 41(2):163–185, 2011.
[3] P. Angelov, X. Gu, Empirical Fuzzy Sets, International Journal of Intelligent Systems, DOI: 10.1002/int.21935, 2017
[4] P. Angelov, X. Gu, J. Principe, D. Kangin, Empirical Data Analysis: A New Tool for Data Analytics, Proc. IEEE Systems, Man and Cybernetics Conference, p.52-59, October 2016; an extended paper to appear in International Journal on Intelligent Systems.

Profile: Prof. Angelov (MEng 1989, PhD 1993, DSc 2015) is a Fellow of the IEEE, of the IET and of the HEA. He is Vice President of the International Neural Networks Society (INNS) and IEEE Distinguished Lecturer (2017-2019). He has 25+ years of professional experience in high level research and holds a Personal Chair in Intelligent Systems at Lancaster University, UK. He formed and led two research groups (Intelligent Systems, 2010-2013 and Data Science, 2014-2017) at the School of Computing and Communications with over 20 academics, researchers and PhD students each and now is the Director of LIRA (Lancaster Intelligent, Robotic and Autonomous systems) Research Centre with over 30 academics. He has authored or co-authored over 300 peer-reviewed publications in leading journals, peer-reviewed conference proceedings, 6 patents, two research monographs (by Wiley, 2012 and Springer, 2002) cited over 7000 times with an h-index of 39 and i10-index of 115. His single most cited paper has 840 citations. He has an active research portfolio in the area of computational intelligence and machine learning and internationally recognised results into online and evolving learning and algorithms for knowledge extraction in the form of human-intelligible fuzzy rule-based systems. Prof. Angelov leads numerous projects (including several multimillion ones) funded by UK research councils, EU, industry, UK MoD. His research was recognised by ‘The Engineer Innovation and Technology 2008 Special Award’ and ‘For outstanding Services’ (2013) by IEEE and INNS. He is also the founding co-Editor-in-Chief of Springer’s journal on Evolving Systems and Associate Editor of several leading international scientific journals, including IEEE Transactions on Cybernetics. He gave over 20 key note talks at high profile conferences. Prof. Angelov was General co-Chair of a number of high profile conferences and a series of annual IEEE Symposia on Evolving and Adaptive Intelligent Systems and more recently on Deep Learning. Dr Angelov is the founding Chair of the Technical Committee on Evolving Intelligent Systems, SMC Society of the IEEE and was previously chairing the Standards Committee of the Computational Intelligent Society of the IEEE (2010-2012). He was also a member of International Program Committee of over 100 international conferences (primarily IEEE).

Prof. Jon Garibaldi

Title: Type-2 Fuzzy Systems for Human Decision Making

Abstract: Type-2 fuzzy sets and systems, including both interval and general type-2 sets, are now firmly established as tools for the fuzzy researcher that may be deployed on a wide range of applications and in a wide set of contexts. However, in many situations the output of type-2 systems are type-reduced and then defuzzified to an interval centroid, which are then often even simply averaged to obtain a single crisp output. Many successful applications of type-2 have been in control contexts, often focussing on reducing the RMSE. This is not taking full advantage of the extra modelling capabilities inherent in type-2 fuzzy sets. In this talk, I will present some of the current research being carried out within the LUCID group at Nottingham, and wider, into type-2 for modelling human reasoning. I will cover approaches and methodologies which make more use of type-2 capabilities, illustrating these with reference to practical applications such as classification of breast cancer tumours, modelling expert variability in cyber-security contexts, and other decision support problems.

Profile:  Professor Jon Garibaldi received the BSc degree in Physics from University of Bristol, UK, in 1984, and MSc degree and PhD degree from the University of Plymouth, UK, in 1990 and 1997, respectively. Prof. Garibaldi is currently Head of School of Computer Science, University of Nottingham, Head of the Intelligent Modelling and Analysis (IMA) Research Group, Member of the Lab for Uncertainty in Data and Decision Making (LUCID) and joint Director of the Advanced Data Analysis Centre (ADAC). His main research interests include modelling uncertainty and variation in human reasoning, and in modelling and interpreting complex data to enable better decision making, particularly in medical domains. Prof. Garibaldi is the current Editor-in-Chief of IEEE Transactions on Fuzzy Systems. He has served regularly in the organising committees and programme committees of a range of leading international conferences and workshops, such as FUZZ-IEEE, WCCI, EURO and PPSN.

Prof. Qiang Shen

Title: Imprecise Data-Driven Feature Selection for Systems Modelling 

Abstract:  Feature selection (FS) addresses the problem of selecting those system descriptors that are most predictive of a given outcome. Unlike other dimensionality reduction methods, with FS the original meaning of the features is preserved. This has found application in tasks that involve datasets containing very large numbers of features that might otherwise be impractical to model and process (e.g., large-scale image analysis, text processing and Web content classification), where feature semantics play an important role.

This talk will focus on the development and application of approximate FS mechanisms based on rough and fuzzy-rough theories. Such techniques provide a means by which imprecisely described data can be effectively reduced without the need for user-supplied information. In particular, fuzzy-rough feature selection (FRFS) works with discrete and real-valued noisy data (or a mixture of both). As such, it is suitable for regression as well as for classification. The only additional information required is the fuzzy partition for each feature, which can be automatically derived from the data. FRFS has been shown to be a powerful technique for semantics-preserving data dimensionality reduction. In introducing the general background of FS, this talk will first cover the rough-set-based approach, before focusing on FRFS and its application to real-world problems. The talk will conclude with an outline of opportunities for further development.
.