Opinion Summarization of Bangla Texts using Cosine Simillarity Based Graph Ranking and Relevance Based Approach

摘要: The main idea of the automatic extractive text or opinion summarization is to find most important representative small subset original document without any loss information. There are many existing methods available for English, Turkish, Arabic and other languages. But very few attempts has been done Bangla language because its having rich morphology multifaceted structure. In this paper, we propose a joint cosine simillarity based graph ranking Relevance scoring approach bangla text. We developed stemming algorithm on Parts Speech(POS) tagging consisting around two lakhs POS tags texts. A redundancy removal also proposed remove so that each sentences in summary represents exactly information document. performance evaluated by measuring recall, precision f-score Rouge metric it showed outperforms