Vector space model is a mathematical model for evaluating the similarities between large data set and a query in increasing order so that a user can find the best matching document among all. It calculates similarity value by using their cosine function. The cosine function evaluates the similarity value by using a weighting scheme. The available factors for weighting schemes are TF(Term-frequency) & IDF(Inverse document frequency). There are various stop words are used when we are writing a query, but only main query terms are important for us for finding best match. It is found that sometimes the results of vector space model are slightly different from other due to the separation of the stop words during similarity analysis. So here we are using some value for stop word so that they can also improve the rank of a document. Also, we are working with entropy-based link optimization algorithm for ranking document, so that we can compare the improved version of vector space model with the entropy-based link optimization algorithm.

Optimization, Entropy, Data Mining