Multi Document Summarization

Abstract

This thesis discusses an sentence extraction approach to multi-document summarization that builds on singledocument summarization methods by using additional, available information about the document set as a whole and the relationships between the documents. Multidocument summarization differs from single in that the issues of compression, speed, redundancy and passage selection are critical in the formation of useful summaries. Our approach addresses these issues by using Agglomerative cluster sentence, GloVe, TextRank, Cosine Similarity. Also this thesis use NLTK as library to filter word such as stopwords, numeric, punctuation, multiple_whitespaces, short_words in order Vectorizing the sentence when using GloVe.