matlab code for quick reduct algorithm in rough set

matlab code for quick reduct algorithm in rough set

Abstract

In order to get a value reduction quickly, this paper puts forwards a new algorithm of value reduction based on attribute-value-tree model in attribute order and proves it's correctness. A attribute reduction and value reduction can be got quickly at the same time from a discrete table in this algorithm. The computational complexity of reduction is changed to O(|U|2|C|) where |U| and |C| are the number of objects and attributes. It is fit to process large data and validated to improve the efficiency by tests.Feature selection refers to the problem of selecting
those input features that are most predictive of a given outcome;a problem encountered in many areas such as machine learning, pattern recognition and image processing. In particular, this has found successful application in tasks that involve datasets containing huge numbers of features which would be impossible to process further. Recent examples include cluster analysis and image classification. Rough set theory has been used as such a
dataset pre processor with much success, but current methods are inadequate at tending minimal reductions. This paper proposes a new feature selection mechanism based on fuzzy forward and backward reduct. It also presents a new entropy- based modification of the original rough set-based approach. These are applied to the problem finding minimal rough set reducts, and evaluated experimentally.

Introduction

Feature Selection (FS) is a process which attempts to select features which are more informative. It is an important step
in knowledge discovery from data. Conventional supervised FS methods evaluate various feature subsets using an
evaluation function or metric to select only those features which are related to the decision classes of the data under
consideration. However, for many data mining applications, decision class labels are often unknown or incomplete, thus
indicating the significance of unsupervised feature selection. However, in unsupervised learning, decision class labels
are not provided. The problem is that not all features are important. Some of the features may be redundant, and others
may be irrelevant and noisy. In this paper, we propose a new rough set-based unsupervised feature selection using
relative dependency measures. The method employs a backward elimination-type search to remove features from the
complete set of original features. As with the classification performance is evaluated using WEKA tool. The method is
compared with an existing supervised method and demonstrates that it can effectively remove redundant features.