Similarity measures are performed by the term frequency- inverse document frequency. TF-IDF can generate the weight vector of Star Student Project set of preference key-words. In our proposed KASR can be implementing in the hadoop map reduce framework for achieving the scalability.