Implementing and Evaluating Collaborative Filtering (CF) Using Clustering

A tremendous increase has taken place in the amount of online content. As a result, by using traditional approaches, service-relevant data becomes too big to be effectively processed. In order to solve this problem, an approach called clustering based collaborative filtering (CF) is proposed in this paper. Its objective is to recommend services collaboratively in the same clusters. It is a very successful approach in such settings where interaction can be done between data analysis and querying. However the large systems which have large data and users, the collaboration are many times delayed due to unrealistic runtimes. The proposed approach works in two stages. First, the services which are available are divided into small clusters for processing and then collaborative filtering algorithm is used in second stage on one of the clusters. It is estimated to decrease the online execution time of collaborative filtering algorithm because the number of the services in a cluster is much less than the entire services available on the web.