Comparitive Study of Vector Space Model Techniques in Information Retrieval for Arabic Language

Abstract

Information retrieval is a process of getting the desired data accurately and efficiently. Arabic Information Retrieval system is considered as an active research area since there are millions Arabic speakers need to retrieve documents, images and videos on the Internet. Arabic is considered as one of the most difficult languages in automatic Natural Language Processing (NLP). All these aspects have encouraged researches to develop Arabic Information Retrieval (IR) systems and models to support the process of retrieving the most relevant documents to the user’s query. In this paper, two techniques of Vector Space Model (VSM) are used to retrieve Arabic text documents which are Cosine Similarity technique and Inner Product technique. The main objective of this research is to compare between these two techniques in terms of retrieving the most relevant documents that met the user’s needs. The result showed that both techniques will retrieve the same order of the documents that are related to the entered user’s query. Both techniques perform well in Arabic information retrieval systems