FSSC: an algorithm for classifying numerical data using fuzzy soft set theory

Introduced is a new algorithm for the classification of numerical data using the theory of fuzzy soft set, named Fuzzy Soft Set Classifier FSSC. The algorithm uses the fuzzy approach in the pre-processing stage to obtain features, and similarity concept in the process of classification. It can be ap...

Full description

Saved in:
Bibliographic Details
Main Authors: Bana Handaga, Bana Handaga, Herawan, Tutut, Mat Deris, Mustafa
Format: Article
Published: IGI Global 2012
Subjects:
Online Access:http://dx.doi.org/10.4018/ijfsa.2012100102
http://dx.doi.org/10.4018/ijfsa.2012100102
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Introduced is a new algorithm for the classification of numerical data using the theory of fuzzy soft set, named Fuzzy Soft Set Classifier FSSC. The algorithm uses the fuzzy approach in the pre-processing stage to obtain features, and similarity concept in the process of classification. It can be applied not only to binary-valued datasets, but also be able to classify the data that consists of real numbers. Comparison tests on seven datasets from UCI Machine Learning Repository have been carried out. It is shown that the proposed algorithm provides better accuracy and higher accuracy as compared to the baseline algorithm using soft set theory.