The paper is devoted to a survey of work done in fuzzy clustering, mainly during the first decade of the 21st century, and that with emphasis on various approachesto the problem, as well as various formulations of the very problem. That is why not only the classical formulations are treated, but several other problems, related to (the use of) clustering, like feature selection, inference systems, three-way clustering, and, on the other hand, such formulations of clustering as the possibilistic one or the one involving intuitionistic fuzzy sets. These are treated as the background for presentation of some specific ideas of the main author, concerning definite heuristic algorithms for effective solving of some of these problems.
The theoretical note deals with the problem of estimation of the value of the least number of objects in fuzzy clusters for following detection of the optimal number of objects in fuzzy clusters through heuristic possibilistic clustering. A technique for detecting the optimal maximal number of elements in the a priori unknown number of fuzzy clusters of the sought clustering structure is reminded and a procedure for finding the initial minimal value of the number of objects in fuzzy clusters is proposed. Numerical examples are considered and conclusions are formulated.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The paper deals with the problem of discovering fuzzy clusters with optimal number of elements in heuristic possibilistic clustering. The relational clustering procedure using a parameter that controls cluster sizes is considered and a technique for detecting the optimal number of elements in fuzzy clusters is proposed. The effectiveness of the proposed technique is illustrated through numerical examples. Experimental results are discussed and some preliminary conclusions are formulated.
The paper deals in the conceptual way with the problem of extracting fuzzy classification rules from the three-way data meant in the sense of Sato and Sato [7]. A novel technique based on a heuristic method of possibilistic clustering is proposed. A description of basic concepts of a heuristic method of possibilistic clustering based on concept of an allotment is provided. A preprocessing technique for three-way data is shown. An extended method of constructing fuzzy classification rules based on clustering results is proposed. An illustrative example of the method’s application to the Sato and Sato’s data [7] is provided. An analysis of the experimental results obtained with some conclusions are given.
The paper deals with the problem of selection of the most informative features. A new effective and efficient heuristic possibilistic clustering algorithm for feature selection is proposed. First, a brief description of basic concepts of the heuristic approach to possibilistic clustering is provided. A technique of initial data preprocessing is described and a fuzzy correlation measure is considered. The new algorithm is described and then illustrated on the well-known Iris data set benchmark and the results obtained are compared with those by using the conventional, well-known and widely employed method of principal component analysis (PCA). Conclusions and suggestions for future research are given.
The interpretability and flexibility of fuzzy classification rules make them a popular basis for fuzzy controllers. Fuzzy control methods constitute a part of the areas of automation and robotics. The paper deals with the method of extracting fuzzy classification rules based on a heuristic method of possibilistic clustering. The description of basic concepts of the heuristic method of possibilistic clustering based on the allotment concept is provided. A general plan of the D-AFC(c)-algorithm is also given. A method of constructing and tuning of fuzzy rules based on clustering results is proposed. An illustrative example of the method's application to the Anderson's Iris data is carried out. An analysis of the experimental results is given and preliminary conclusions are formulated.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.