PERFORMANCE EVALUATION OF INFORMATION RETRIEVAL SYSTEM USING VECTOR SPACE MODEL: A COMPARATIVE ANALYSIS

Authors

  • Omar Al-rassam Koya University
  • Miran Hama Saeed Mohammed Amin Koya University
  • Zhenar Shaho Faeq Koya University

DOI:

https://doi.org/10.25195/ijci.v47i2.332

Keywords:

Information Retrieval, Vector space model, inverse document frequency, Term frequency, stemming

Abstract

The increasing use of the internet has created a vast amount of digital information and it is expanding extremely fast. Therefore, Information retrieval becomes a challenging task to fetch relevant information for users. The aim of this paper was to examine and evaluate the performance of the Information retrieval system through eight experiments to test all the features that can be used in a vector space model. These experiments were compared to show the best and the worst implemented features. The features are represented by applying (tf.idf, stop words, stemming), (tf.idf, No- stop words, stemming), (tf.idf, No- stop words, No-stemming), (tf.idf, stop words, No-stemming), (tf, stop words, stemming), (tf, No- stop words, stemming), (tf, No- stop words, No-stemming), (tf, stop words, No-stemming). Results showed that using stop words, stemming approach, and tf.idf improve the performance of the system. However, when tf was used without using stop words and stemming approaches the performance of the system is declined. In addition, results showed that stop words have a significant effect on the system while the stemming approach has no noticeable effect particularly with tf.

Downloads

Download data is not yet available.

Author Biographies

Omar Al-rassam, Koya University

Department of Mathematics, Faculty of Science and Health

Miran Hama Saeed Mohammed Amin, Koya University

Department of Software Engineering, Faculty of Engineering

Zhenar Shaho Faeq, Koya University

Department of Software Engineering, Faculty of Engineering

Downloads

Published

2021-09-07