Text Categorization using Support Vector machines

Abstract

The categorization of texts is an application that falls within the discipline of naturallanguage pro­ cessing and is closely related to the concept of classification. Due to the abundant existing infor­ mation becomes necessary to organize, maintain, and process any information available from a deeper knowledge of the language of the support vector machines (MSV) belong to the family of linear classifiers, and can be used to resolve the problem of the categorization of texts(eT) which consists in label text or document with one or severa! predefined thematic categories. The reason which tackles the problem is their application in different scenarios of the area of information retrie­ val (IR) such as the automatic organization of documents, filtering of documents. The approach of the MSV basically considers the following: Given a set of documents D anda set of categories e, it is important to find a function that match toa document d taken from D, a particular category e in C.