Comparison between vector space model and fuzzy set model retrieval of arabic collection

Information Retrieval Systems should provide the users with more relevant documents for their queries. There are many models for calculating the similarity between a query and the documents. We used Vector Space model using cosine similarity and Fuzzy Model then eva-luated the results for each on an Arabic collection. We used the inverted file for indexing, stemming the Arabic words using suffix, infix and prefix removal. We observed that Fuzzy can give higher recall than Vector Space but in general Vector Space gives better precision and it depend on the threshold value. We used MS visual basic to implement our project and connected it with MS Access which we used to build the data base for our work.