Myanmar News Retrieval in Vector Space Model using Cosine Similarity Measure

Abstract

Information Retrieval (IR) is an effective means of retrieving the best relevant document to user query. Nowadays, the problem of documents similarity deals with IR is retrieving required information from a large amount of data. In this paper we studied a Vector Space Model (VSM) that is used in IR and represents a document as a vector in an n-dimensional space, where each dimensional represents a term and measured through cosine angle between two vectors. The objective of this paper is to retrieve the relevant file from Myanmar news data sets using cosine similarity measure in VSM to user's query. Evaluations are done in terms of similarity score by Precision, Recall and F-score.