Review on Service Recommendation System using Social User?s Rating Behaviors

Abstract

The research communities of information retrieval, machine learning and data mining are recently started to paying attention towards Service recommendation systems. Traditional service recommendation algorithms are often based on batch machine learning methods which are having certain critical limitations, e.g., mostly systems are so costly also new user needs to pay the certain cost for new login, can?t capture the changes of user preferences over time. So that to overcome from that problem it is important to make service recommendation system more flexible for real world online applications where data arrives sequentially and user preferences may change randomly and dynamically. The proposed system present a new framework of online social recommendation on the basis of online graph regularized user preference learning (OGRPL), which incorporates both collaborative user-services relationship as well as service content features into an unified preference learning process. Also provide aggregated services in only one application (social networking) which increases user?s interest towards the services. Proposed system also provides security about subscribed services as well as documents/photos on online social network application. Proposed system utilizes services like Active Life, Beauty & Spas, Home Services, Hotels & Travel, Pets, Restaurants and Shopping.