Improving Feature Vector for Extractive Text Summarization of Legal Judgement

Extractive Text summarization in legal judgment has the key task of finding important information from the legal text and representing it in concise summaries. Large amounts of legal documents can be effectively produced into summaries through the use of extractive text summarization. Finding the most important and informative details for summarizing might be difficult, though, due to the volume of material in legal papers. The task of feature selection in legal documents for extractive text summarization is covered in this article. The goal of this study is to examine and evaluate the value of feature selection methods for enhancing the accuracy and efficiency of legal text summaries. Features like sentence length, sentence position, Term frequency and inverse sentence frequency (TF-ISF), degree of similarity, keywords, and citation are used. To extract the most important details from legal papers, this paper provides a methodology that incorporates both generic text elements and domain-specific legal features. This feature selection method will enable us to find the most important sentences in the judgment. Using a large Indian-Extractive dataset (IN-Ext) of legal documents, the proposed approach is examined.