Silt?�?E Text Information Retrieval System Based On Vector Space Model

Abstract

The main aim of an information retrieval system is to extract appropriate information from enormous collection of data based on user?�?s need. In this research, we were concerned on text information retrieval for Silt?�?e language. We discussed its basic morphological features that have effect on the implementation and performance of information retrieval systems. In detail, we discussed about the model we used, that is the Vector Space Model, the architecture, prototype development tool and evaluation of the developed prototype. Different text preprocessors are offered by Apache Solr. For our prototype implementation we used the Apache Solr?�?s standard tokenizer, standard stopword removal algorithms and term weighting and similarity computing functions. The prototype is evaluated using the corpus that was prepared from Silt?�?e school text books. We used ranked retrieval evaluation techniques which are average precision and mean average precision. The experimental analysis shows that the obtained mean average precision 81.4% result is encouraging to develop vertical Silt?�?e text information retrieval systems and Silt?�?e search engines.