Restricted Legal Dictionary generation for Legal Domain Specific Mining

Abstract

The Indian legal domain maintains largest heterogeneous multi domain based text corpora. In order to document specific search domain specific dictionaries are widely acceptable compared to whole legal dictionary. Many domain specific dictionary extractions proposed by various authors for business and scientific domains. In this paper we proposed a two stage domain specific dictionary for legal domain. The approach helps to construct Restricted Legal Dictionary which is domain specific in nature. The dictionary constructed supports text mining mechanisms over specific crime (dowry) based legal documents and improves the BoW construction with higher domain specific terminology. The Restricted Legal Dictionary constructed in this paper improves the efficiency of similarity measures during document comparison techniques