The Study Of Point-Of-Interest Recommendation In Location-Based Social Network

Recently, the social network is becoming widely available and useful tools for communication between users. Every day, targeted audiences furnish content of different information, from multimedia data to privacy of personal information. Using social network can find any information about any Individual, his habits or interests and any basic general required. Consequently, the popularity of social network is increasing vastly and is becoming boundless. Evolution within social network developed a new kind of internet communication and more researches gave a new branch in the area of internet communication. During 2001, a location- based social network emerged. In other words, a social structure made up of individuals connected by one or more specific types of interdependency, such as friendship, common interests, and shared knowledge. Generally, a social networking service builds on and reflects the real-life social networks among people through online platforms such as a website, providing ways for users to share ideas, activities, events, and interests over the Internet. A location can be represented in latitude, longitude coordinates, relative(100 meters north of the Space Needle), and symbolic(food, shop, or education) form. A location usually has three kinds of geospatial representations, consisting of a point location, a region, and a trajectory. In other words, LBSN is a type of information and entertainment services, based on current location, sensing users that take advantage of mobile devices.These location-based social networks orient not only on person location, but it describes person's POI(Point-of-Interests). LBSN specializes in user comments that include a review of visited places, previous check-ins and likes or dislikes. Which means that LBSN allows us to analyze our own personal life and represent them as an algorithm. There are sequences of a person's habits, which can be represented as a personal life model.Usually, one's personal life model repeats itself on the specific location, because he/she chooses those places that are convenient to him. In the new location, a great amount of POI bewilders the person and complicates his decision making. Recommendation system allows making the decision much simpler, using objects(films, music's, books, news, websites) that interest the user, depending on the information on his profile. Often recommendation system makes decision on a single object, but it is research oriented on a set of POI depending on a foundation of person life model. Basically, I was recommended on a changing person's location using the set of POI but I that personalized life model is more precise.In this paper, I am focusing on Location Based Social Network and its capabilities. From the aspect of the regularity of user's movements and describe user's activities, i.e. location's category information, to recommend personalized life model to a certain user which matches the regularity of his/her activity.The study POI Recommendation in LBSN, as to analyze the personal habits, behaviors and their daily cycle in the specific location, then recommend them to adapt quickly in the new location according to his/her order preference.Considering location semantics, to use both location's geographical information and semantic information, to better understand user's movements, process user's raw trajectory and extract location trajectories on geographical space and POI trajectories to describe user's movement, and get user's experiences under different location categories, thus, requiring us to build user model. Personal Life Model in this research, calculates user's similarity, extract candidate's visited locations which match user's habits(life model) from similar users, through scoring strategy, which considers the information of user's own life model, its entire similarities within one category, those similarities under different categories and location's popularity under its own category, to recommend top-k locations is according to one's person's life model.The experiment was conducted by Foursquare data set, which is most popular LBSN in the world and currently has more than 8 million registered users. Experimental results show that every third user have own life model, making this experiment much meaningful. The study of POI recommendation has purpose and scalability because recommendation of LBSN data is not limited and requires further study, which is the expansion of this topic.