Gabrys, B., 2002. Neuro-fuzzy approach to processing inputs with missing values in pattern recognition problems. International Journal of Approximate Reasoning, 30 (3), pp. 149-179.
Full text not available from this repository.
Official URL: http://www.sciencedirect.com/science?_ob=ArticleUR...
An approach to dealing with missing data, both during the design and normal operation of a neuro-fuzzy classifier is presented in this paper. Missing values are processed within a general fuzzy min–max neural network architecture utilising hyperbox fuzzy sets as input data cluster prototypes. An emphasis is put on ways of quantifying the uncertainty which missing data might have caused. This takes a form of classification procedure whose primary objective is the reduction of a number of viable alternatives rather than attempting to produce one winning class without supporting evidence. If required, the ways of selecting the most probable class among the viable alternatives found during the primary classification step, which are based on utilising the data frequency information, are also proposed. The reliability of the classification and the completeness of information is communicated by producing upper and lower classification membership values similar in essence to plausibility and belief measures to be found in the theory of evidence or possibility and necessity values to be found in the fuzzy sets theory. Similarities and differences between the proposed method and various fuzzy, neuro-fuzzy and probabilistic algorithms are also discussed. A number of simulation results for well-known data sets are provided in order to illustrate the properties and performance of the proposed approach.
|Additional Information:||This paper is one of the series of papers leading towards a development of a general framework for adaptable classification systems utilising General Fuzzy Min Max algorithms which we have developed over a number of years. The novel contribution of the paper is concerned with a different way of dealing with missing values not requiring imputation of unknown values. This approach also facilitates dealing with variable number of input dimensions without any changes to the existing learning algorithms which is a highly desirable feature not present in any of the comparable learning algorithms. The paper had been listed in the Top 25 Hottest Articles of this very well known IJAR journal between 2002 and 2004 - an official list of the most often downloaded articles online kept by ScienceDirect.|
|Subjects:||Generalities > Computer Science and Informatics > Artificial Intelligence|
Generalities > Computer Science and Informatics
|Group:||School of Design, Engineering & Computing > Smart Technology Research Centre|
|Deposited By:||INVALID USER|
|Deposited On:||17 Dec 2007|
|Last Modified:||07 Mar 2013 14:35|
|Repository Staff Only -|
|BU Staff Only -|
|Help Guide -||Editing Your Items in BURO|