Fuzzy Sets in Human-Centric Systems

By Witold Pedrycz Introductory comments and selected areas of applications. Developments in human-centric systems. Fuzzy sets as enabling technology. Human-centricity of systems and a role of fuzzy sets. Selected case studies in pattern recognition, system modeling, data analysis, decision-making, software engineering, web engineering Fuzzy sets and information granulation: basic concepts, terminology and motivation. Design of fuzzy sets: data and user-based approaches. Selected algorithms. Logic operations on fuzzy sets: t-norm s and t-conorms, uninorms; design of logic operators Transformations of fuzzy sets and fuzzy relations Higher order constructs: interval-valued fuzzy sets, type-2 fuzzy sets Fuzzy models and granular models: design methodology, development algorithms, interpretability accuracy tradeoffs, validation and verification. Rulebased architectures Fuzzy clustering and unsupervised pattern recognition. Data-driven algorithms and mechanisms of partial supervision. Proximity-based clustering. Distributed clustering. Interpretation and reconstruction issues. Fuzzy neural networks: Taxonomy of logic neurons and their characteristics. Heterogeneous topologies of the networks. Fuzzy neural networks as pattern classifiers. Interpretation of networks and knowledge-directed learning s and t-conorms, uninorms; design of logic operators Transformations of fuzzy sets and fuzzy relations Higher order constructs: interval-valued fuzzy sets, type-2 fuzzy sets Fuzzy models and granular models: design methodology, development algorithms, interpretability accuracy tradeoffs, validation and verification. Rulebased architectures Fuzzy clustering and unsupervised pattern recognition. Data-driven algorithms and mechanisms of partial supervision. Proximity-based clustering. Distributed clustering. Interpretation and reconstruction issues. Fuzzy neural networks: Taxonomy of logic neurons and their characteristics. Heterogeneous topologies of the networks. Fuzzy neural networks as pattern classifiers. Interpretation of networks and knowledge-directed learning