Problem Definition, goal and Importance:

Stemming is one of the methods that can be applied in information retrieval systems in order to increase the quality of the results in both recall and precision. Its effect especially on recall measure can be interesting, since the range of retrieved documents would increase .In this project we observed the influence that stemming would have on vector space IR engines in Farsi. We followed three main steps in fulfilling our goals. First we had to apply the stemming process to our collection which was Hamshahri collection. In the second stage we needed to run the vector space engine on the new collection. And in the final phase the comparison between the results gained from stemmed collection and the collection without stemming was made. We will look at each step in detail in the next sections.