IFSA - NAFIPS 2019

2019 IFSA World Congress and NAFIPS Annual Conference

June 18-21, 2019 - Lafayette, Louisiana, USA

[Return to main page]

IFSA - NAFIPS 2019: Special Sessions

Fuzzy Systems as Machine Learning Algorithms

Organizer: Barnabas Bede, DigiPen Institute of Technology, Redmond, Washington, USA

Emerging Topics in Computing Paradigms and Approximate Computing

Organizer: M. Hassan Najafi, University of Louisiana at Lafayette, USA

Fuzzy Optimization and Decision Making: Theory, Algorithms and Applications

Organizers: Weldon Lodwick, University of Colorado, Denver, USA and
Masahiro Inuiguchi, Osaka University, Japan

This special session concentrates on the areas of fuzzy and possibilistic optimization and decision-making. Topics for this special session include:

  • Fuzzy Optimization, static and dynamic
  • Possibilistic and Interval Optimization, static and dynamic
  • Optimization under Uncertainty
  • Decision Making
  • Fuzzy Analytical Hierarchy Processes
  • Applications in fuzzy and possibilistic optimization and decision making and analytical hierarchy processes

Similarity, Correlation and Association Measures

Organizers: Ildar Batyrshin, Centro de Investigación en Computación (CIC), Instituto Politécnico Nacional (IPN), CDMX, México and
Vladik Kreinovich, University of Texas El Paso, USA

Similarity, correlation and association measures as measures of relationship between data are studied in statistics, information retrieval, data mining and data science, classification and machine learning, recommender systems and decision-making. They have applications in ecology, social and behavioral sciences, biology and bioinformatics, social network and time series analysis, image, video and natural language processing. These measures have been introduced on different domains: on the set of variables, images, time series, vectors of attributes, probabilistic distributions, fuzzy sets of different types etc. Recently, the similarity, dissimilarity and correlation measures are studied as functions defined on universal set (domain) and satisfying some simple sets of properties. Generally, any function of two or more arguments having reasonable interpretation can be considered as an association measure establishing relationship between them.

Topics for this special session include:

  • New similarity, correlation, interestingness, agreement, dependence and other association, relationship measures defined on specific domains;
  • Association measures as functions defined on universal set, the properties of such functions;
  • Association functions and fuzzy (valued) relations;
  • Methods of construction of association measures;
  • Transformations of association measures;
  • Transformation of data used in association measures;
  • Relationships between different types of association measures;
  • Visualization of association measures;
  • Statistical methods of analysis of new association measures;
  • Applications of similarity, correlation and association measures.

Fuzzy Systems in Economics and Management

Organizers: José M. Merigó, University of Technology Sydney, Australia, and University of Chile and
Jose M. Brotons, Miguel Hernandez University, Spain

Economics and management are social sciences fundamental for our modern society. The development of quantitative theories for a better understanding of these fields has grown a lot during the last years. Particularly, there are many new contributions appearing in the literature that use fuzzy systems and other computational intelligence techniques for obtaining a better representation of a business or economic problem. This Special Session aims to present applications of fuzzy systems and computational intelligence in any area connected to economics and management including decision making, finance, marketing, accounting, strategy and econometrics. This session is sponsored by the International Association for Fuzzy Systems in Management and Economics (SIGEF) (http://www.sigef.net/).

Topics and areas for this special session include:

  • Fuzzy economics
  • fuzzy management
  • fuzzy business
  • fuzzy finance
  • fuzzy marketing
  • fuzzy accounting
  • fuzzy strategy
  • fuzzy econometrics
  • fuzzy insurance
  • decision making
  • expert systems
  • other computational intelligence applications in economics and management

Fuzzy Multisets: Mathematical Foundations and Applications

Organizers: Susana Montes, University of Oviedo, Spain and
Irene Díaz, University of Oviedo, Spain

Fuzzy multisets are a generalisation of fuzzy sets whose membership value is a multiset in the [0, 1] real interval. This generalisation is related to other generalizations such as to hesitant fuzzy sets. However fuzzy multisets allow repetition of the individual membership values, representing a new paradigm. This need to account for repeated membership values has been recognised in the literature about hesitant fuzzy sets. But despite the similarities, we cannot regard the typical hesitant fuzzy sets as a particular case of the fuzzy multisets and neither can we identify the fuzzy multisets with the multiset-based hesitant fuzzy sets. This special session welcomes both theoretical and application papers focused on developing the mathematical foundations of fuzzy multisets or on showing the (di)similarities between these sets and other generalisations. Papers focused on the application of fuzzy multisets to a real domain are also welcome.

Inter-Relation Between Interval and Fuzzy Techniques

Organizers: Martine Ceberio, University of Texas El Paso, USA and
Vladik Kreinovich, University of Texas El Paso, USA

The relation between fuzzy and interval techniques is well known; e.g., due to the fact that a fuzzy number can be represented as a nested family of intervals (alpha-cuts), level-by-level interval techniques are often used to process fuzzy data. At present, researchers in fuzzy data processing mainly used interval techniques originally designed for non- fuzzy applications, techniques which are often taken from textbooks and are, therefore, already outperformed by more recent and more efficient methods.

One of the main objectives of this special session is to make the fuzzy community at-large better acquainted with the latest, most efficient interval techniques, especially with techniques specifically developed for solving fuzzy-related problems. Another objective is to combine fuzzy and interval techniques, so that we will be able to use the combined techniques in (frequent) practical situations where both types of uncertainty are present: for example, when some quantities are known with interval uncertainty (e.g., coming from measurements), while other quantities are known with fuzzy uncertainty (coming from expert estimates).

Topics and areas for this special session include:

  • interval computations, especially topic of potential and actual interest to fuzzy community
  • interval uncertainty
  • interval-valued fuzzy sets

Aggregation and pre-aggregation functions. Theory and Applications

Organizers: Humberto Bustince, Universidad Pública de Navarra, Spain,
Daniel Paternain, Universidad Pública de Navarra, Spain,
Javier Montero, Universidad Complutense de Madrid, Spain, and
Radko Mesiar

Information fusion plays a crucial role in many real-world applications. Formally, aggregation functions and its recent generalizations, such as pre-aggregation functions, are one of the most important mathematical tools to deal with information fusion. In this special session we intend to cover topics related with aggregation and pre-aggregation functions both from a theoretical and an applied point of view. If we focus on aggregation functions, we invite recent developments in classical functions, such as t-norms, t-conorms and means, with a special focus on fuzzy measure-based integrals, such as Choquet or Sugeno integrals, and new families of aggregation functions, such as overlap or new penalty-based functions. Moreover, due to the development of new forms of monotonicity (weak monotonicty, directional monotonity or ordered directional monotonicity), a new broad family of fusion functions have appeared in the literature, encompassing classical aggregation functions but opening a new research direction. Finally, from an applied point of view, we intend to discuss novel applications in which the use of aggregation or pre-aggregation functions have demonstrated to improve significantly the quality of previous results. Some of these applications could be classification or regression problems, image processing, neural networks and deep learning, among other.

Granular Computing

Organizers: Shusaku Tsumoto, Shimane University, Japan, and
T.Y. Lin, San Jose State University, USA

Granular Computing (GrC) is an emerging computational and mathematical theories that involve the concept of granules. A granule can be a sub-Turing machine, a piece of elementary knowledge, or a region of uncertainty, which  can be viewed as elements of human intelligence.  Though the label is  relatively recent, the basic notions and principles, though under different  names, have appeared in many related fields, such as neighborhood systems  and infinitesimals in the foundation of granular computing, divide  and conquer in theoretical computer science, information hiding in software  engineering, interval computing, fuzzy and rough set theories  probability/possibility/belief measures in uncertainty mathematics,  granularity in artificial intelligence, neutrosophic computing, quotient space  theory, machine learning, databases, and many others. This special session  will give a research forum in which researchers on cognitive computing,  intelligent computing and granular computing can exchange their ideas.

Speakers:

  • T.Y.Lin: Differentiable Zadeh Sets: towards a theory for Fuzzy Control
  • Dominik Slezak: Scaling out "RoughSets" - a fuzzy-rough package in R
  • Shusaku Tsumoto: Statistical and Fuzzy Mixture Models for Sample Decomposition
  • S.L. Wang and T.P. Hong: Granular Data Mining
  • Ying Xie : Synthesizing a fuzzy kernel function for K-NN classification via deep learning