The Research On Information Recommendation Of Location Based On Social Network

In recent years,people have become to rely on the intelligent mobile terminal for accessing and sharing information in real-time.As the development of the global positioning system and communication technology,the location based on social network is becoming popular with users gradually.The abundant users and projects resources becomes the primary channel for users to look for projects' information online,and numbers of users lie behind a vast and rich geographical location information.At the same time,it causes the“information overload”problem.Nowadays,using this kind of information to extract users' information and recommend specific preference information has become the urgent and important problem to be solved.This article will summarize the current and relevant research achievements and consider the multi-source and heterogeneous information.Also,it designs the personal recommendation model based on transmission mechanism on the basis of multi-source and heterogeneous information.The purpose of this thesis is to realize the most and best utility and benefit for the users,businesses,service providers and other stakeholders,also optimize and innovate the mode of personal service.The thesis regards the geographic information as research object based on location based on social network and focuses on the model of personalized information recommendation research in network information environment on the basis of information rule.First of all,the paper introduces the application and academical background,then determines the research content and the research train of thought,also clears out the emphasis and difficulty of the research.During the analysis,it reaches that informationtransmission mechanism is the basis of the personal recommendation system.In this paper,it draws the conclusion that users tend to spread preference information after the research of the geographic information dissemination mechanism and the simulation method of quantifying geographic information dissemination process.Furthermore,it uses the "accumulation effect" of linear threshold model to analyze the influential factors of user preference.Secondly,this thesis models multi-source heterogeneous geographic information by the aid of vector space model and similarity calculation method.Then it uses bayesian theory to study the user preferences.Finally,this thesis uses the classical model of collaborative filtering recommendation and mixes the multi-source heterogeneous geographic information to design recommendation algorithm based on the users and projects respectively.At last,the thesis uses three recommendation result evaluation indexes:the root mean square error,the accuracy and coverage on the Foursquare data sets.The experimental results show that the modified recommendation algorithm is better in line with the actual results of user preferences compared with the traditional recommendation algorithm.To a certain extent,the modified model optimizes the personalized information recommendation and solves "information overload" problem under the Location Based on Social Network.The research results show that by understanding the information dissemination mechanism firstly combined with information recommendation research could better understand the information propagation law under Location Based on Social Network,then,the thesis lays on a good theoretical basis on the personal information recommendation pattern.