Main Article Content
Abstract
The social networks are build to share the user reviews, photos, videos and opinions on products and locations. The geographical location information are collected with smart phone based applications. The check in and user posting details are maintained under the Location Based Social Networks (LBSN). The huge volume of data values are processed with the big data environment. The big data values are processed with the support of the cloud resources. The user spatial parameters and review information are utilized in the service rating prediction operations. The Location Based Rating Prediction (LBRP) algorithm is applied for the service rating prediction process. The LBRP uses the three measurements for the prediction process. The useritem geographical connections, user-user geographical connections and interpersonal interest similarity measures are estimated for the rating prediction process. The user – item geographical connection indicates the distance between the user ratings and the user – item geographical locations. The user – user geographical connection indicates the user rating differences and the user – user geographical location distances. The optimal service discovery operations are carried out with user behavior discovery and service score index methods. The user behavior estimation process is carried out with multi activity centers. The service score estimation and index operations are performed with the attributes of the Point of Interests (POI). The user recommendation task is integrated with the system to suggest better services with reference to the user behavior and ratings. The user category, region and seasonal aspects are employed in the multi criteria based recommendation process.