Community based Personalized Location Recommendation System

This paper introduces Community Based Personalized Location Recommendation System (CMP) framework that provides the user based on their personal preferences with the most interesting location. The framework divided into two parts. The first part identifies his/her relevant community based on user-community similarity. Then from the identified community and dedicated sub-community location, this part recommends interested locations based on user-location similarity for each user. The second part develops communities for users within the user-specified preference based on their categories; communities that represent user-preferences categories and will recommend based preference locations. That improves recommendation on Location Based Social Networks (LBSNs) and takes into consideration both efficiency and quality, it also makes the system easily solve the new user problem and handling the data sparseness problem for location recommendations. Spatial Group analysis cluster technique, it discovers community location shape that divides location into the group based (X, Y) coordinates. It means dividing the city into sub-community that increase real-time location and facilities to dedicate location performance and efficiency to generate candidate location and local expert user that aid to increase recommendation process. The results of our framework achieve good quality, high performance and compared with existing approaches.