Research On Personalized POI Group Recommendation Algorithm Based On LBSN

With the popularity of the Internet in the world,the Internet has become the most important bridge for people to communicate with each other and obtain information.In order to meet people's need for information exchange,social networks have emerged.In order to continuously improve the user's relevant experience,and enable users to efficiently find content that suits their interest preferences in the presence of massive amounts of network information,the personalized recommendation system was studied by relevant scholars.However,in the current location-based social network(Location-Based Social Network),most of the personalized recommendation algorithms are biased toward the user's single point of recommendation,that is,there is no connection between the various recommended POI(Point of Interest)points,or just a simple sorting.This recommendation has spatial locality,and it cannot satisfy the changing needs of users.The main content of this paper's research is the POI group recommendation algorithm,which focuses more on the relevance of the recommendation results.It mainly includes the following four aspects:First of all,this paper proposes a scheme for mining hotspots in cities and potential hotspots for users.By analyzing the historical check-in data of users in the city and finding the hotspots in the city,it is extremely important and useful for the planning and construction of the city.It is convenient for city planners to master the development of the city from an overall perspective.The acquisition of the user's potential living area can help the target user to more accurately and efficiently find the location range and location information he likes.Secondly,this paper proposes the target user's local living model and the remote living model to obtain the program.Through the use of association rules and topic models and other related technologies,the user's life pattern is obtained,and a set of POI type transfer sequences that meet the user's living habits is obtained,making the recommendation The results are more relevant and join the local user's life to ensure the flexibility and accuracy of off-site recommendations.Afterwards,this paper proposes a solution for the user's personalized POI group,combining the above two aspects,after the user's potential life hotspot area and the user's life mode are obtained,by mapping the POI category to the specific POI locationin the user's potential life hot spot area.In the above,the obtained POI type transfer sequence set is converted into a POI position transfer sequence set,and the final recommendation result is obtained and recommended to the user.This article has verified the recommendation results and got a good recommendation accuracy and uses some of the user data from the Four Square dataset in California.Finally,this paper builds a relevant display platform of the recommendation system,and displays the information of the city POI information,the user's potential living area,and the user's personalized recommendation results on the page.