Improved web service recommendation via exploiting location and QoS information

Web services describe a way of integrating web-based applications that help in machine-to-machine interaction over the network. There are many publicly available web services and the number keeps on increasing. However this ever increasing pool makes it difficult for optimal service selection. So, appropriate selection of web service suiting the requirements of user is a non-trivial task. Our research proposes technique which helps in optimal service selection with optimal Quality-of-Service (QoS) performance. The technique designs a recommender system based on Collaborative Filtering (CF) algorithm which employs location and QoS values to cluster users and services. The main objective of the proposed technique is to address the issue of data sparsity and scalability. The proposed approach uses k-Nearest Neighbor (kNN) algorithm with Support Vector Machine (SVM) in CF algorithm framework. SVM a state-of-the-art classification algorithm used to address the issue of sparse data and k-NN used with CF algorithm for similarity mapping of user and services.