Online communities will provide the trustworthiness of their services and also, recommendation systems to improve the commercial value in this competitive business world. Prediction is the greatest method to get people interested whatever offered. Traditional QoS based prediction approach, predicts the QoS value of web service when the incompletion QoS records. This proposed approach introduced cluster based PSO algorithm, which provides better scalability, simplicity, and efficiency. It uses the density-based clusters based on web service users' location and ranks the web services based on PSO algorithm. Here, top-K users are selecting based on web service preferences and weights are giving for experienced neighbors. To achieve the high-quality outcome of the ranking sequence by the control of fitness function and verified by AP correlation coefficient method. The experimental results discussed how this proposed approach provided better prediction accuracy and compared with other existing approaches.
