Towards a Unified Modeling and Knowledge-Representation Based on Lattice Theory: Computational Intelligence and Soft Computing Applications

ISBN-13: 9783642070587 / Angielski / Miękka / 2010 / 245 str.

Vassilis G. Kaburlasos

Opis

This research monograph proposes a unified, cross-fertilizing approach for knowledge-representation and modeling based on lattice theory. The emphasis is on clustering, classification, and regression applications. It presents novel tools and useful perspectives for effective pattern classification. The material is multi-disciplinary based on on-going research published in major scientific journals and conferences.

Spis treści

The Context.- Origins in Context.- Relevant Literature Review.- Theory and Algorithms.- Novel Mathematical Background.- Real-World Grounding.- Knowledge Representation.- The Modeling Problem and its Formulation.- Algorithms for Clustering, Classification, and Regression.- Applications and Comparisons.- Numeric Data Applications.- Nonnumeric Data Applications.- Connections with Established Paradigms.- Conclusion.- Implementation Issues.- Discussion.

Blurb

By ‘model’ we mean a mathematical description of a world aspect. With the proliferation of computers a variety of modeling paradigms emerged under computational intelligence and soft computing. An advancing technology is currently fragmented due, as well, to the need to cope with different types of data in different application domains. This research monograph proposes a unified, cross-fertilizing approach for knowledge-representation and modeling based on lattice theory. The emphasis is on clustering, classification, and regression applications. It is shown how rigorous analysis and design can be pursued in soft computing using conventional (hard computing) methods. Moreover, non-Turing computation can be pursued. The material here is multi-disciplinary based on our on-going research published in major scientific journals and conferences. Experimental results by various algorithms are demonstrated extensively. Relevant work by other authors is also presented both extensively and comparatively.