Research On Personalized Service Of Mobile Social Network Based On Trust Assessment

In the current mobile environment,traditional PC services and applications in the industry have moved to the mobile field.The mobile social network makes full use of the convenience brought by the technological change,and supports the users' needs at anytime and anywhere,and has quickly become the most hot and potential mobile network application.With the growing trend of diversification in all walks of life,mobile social network has gradually become one of the main platforms for people to obtain information and services in daily life.However,the increasing number of users and the low cost of information delivery in the network make users in mobile social networks face serious information overload problems,The miscellaneous and loose trust content make it difficult for users to accurately and quickly obtain services that meet their needs.In addition,due to the inherent virtual nature of the network,the information in the mobile Internet still has security problems such as the insurability of information quality,the difficulty of information service publishers monitoring,and malicious fraud.Considering the possibility of property and related losses,there is considerable trust concern between users and providers in mobile social networking.In view of the above problems,this paper proposes a solution based on the trust evaluation of mobile social network personalized service recommendation.By evaluating and quantifying the user's trust network in the mobile social network,and then providing support for personalized service recommendation,the problems of overloading the mobile social network user information are effectively solved,and the bothering users caused by fake and bad information is also effectively avoided.In addition,in order to better provide personalized service for mobile social network users,improve service quality and enhance user experience,this article starts from the two characteristics(mobility and sociality)of mobile social network,introduces the user's geographical location information,Constructing a new dimension of user characteristics and using the geographical similarity between users as the signal input for the personalized service recommendation model to achieve the purpose of optimizing the recommended results.In the algorithm selection of recommendation model,because of the advantages and disadvantages of service recommendation based on collaborative filtering and content filtering,this paper will adopt a hybrid recommendation model based on collaborative filtering and content-based filtering,Making personalized recommendation of the service better.Finally,using the TRA(rational behavior theory),TPB(planned behavior theory),and TAM(technical acceptance model)which are widely accepted in the academic field to evaluate the user's willingness to use the recommended list generated by the recommendation model.We can evaluate users' willingness to adopt services and predict potential behaviors of users through user behavior perception.The user's willingness to use will be able to reflect the user's value requirements fairly,and it can better judge the demand relationship between the users and the recommended results in the mobile social network based on the assessment of the willingness to use,and re-input it to the personalized service recommendation model as a feedback signal,making the re output recommendation service more matching the user's needs.