PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Feature Selection for Dimensionality Reduction
Dunja Mladenić
In: Subspace, latent structure and feature selection : statistical and optimization perspectives workshop, SLSFS Lecture notes in computer science , 3940 . (2006) Springer , Berlin; Heidelberg , pp. 84-102.

Abstract

Dimensionality reduction is a commonly used step in machine learning, especially when dealing with a high dimensional space of features. The original feature space is mapped onto a new, reduced dimensionally space. The dimensionality reduction is usually performed either by selecting a subset of the original dimensions or/and by constructing new dimensions. This paper deals with feature subset selection for dimensionality reduction in machine learning. We provide a brief overview of the feature subset selection techniques that are commonly used in machine learning. Detailed description is provided for feature subset selection as commonly used on text data. For illustration, we show performance of several methods on document categorization of real-world data.

EPrint Type:Book Section
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Information Retrieval & Textual Information Access
ID Code:2329
Deposited By:Dunja Mladenić
Deposited On:22 November 2006