Researches On Mobile Personalized Information Recommendation Service Oriented User Private Concern Problem

Along with the rapid development of mobile intelligent terminal and wireless network technology,users have growing demand to easily access the Internet information service whenever and wherever.Mobile commerce application is constantly breaking through in deeper and broader dimensions.The mobile personalized information service and recommendation system is a key application in mobile commerce that can effectively relieve the information overload and improve quality of customer service.Its application level has become an important index to measure the innovation and development ability of Internet enterprise.User is enjoying the high quality of mobile personalized commerce service but it comes with private concern problem.In the meantime,personality traits of mobile user,social relations,emotion tendency,context and other factors have an important influence for user's network preference.The high quality mobile personalized information recommendation service under privacy concerns becomes very challenging.Therefore,how to utilize the personalized recommendation technology to improve user adoption mobile personalized information recommendation service will,and to provide a high quality of recommendation service but reduce users' privacy concerns is imminent.This study explores psychological behavior proceduce of mobile users' privacy concerns.Paper focused on factors that affect users'privacy concerns whose degree affect adoption mobile personalized information recommendation service.On this basis,Paper researches the theory and method of mobile personalized information recommendation oriented user privacy concerns.It takes theory of reasoned action,social interactions,psychological behavior and privacy concerns as theoretical basis.It adopts text mining,context compute,SEM,collaborative filtering recommendation method to realize high quality recommendation service and reduce the degree of adoption intention of mobile personalized information recommendation service.The main research work is as follows:1.Research influential factors of users' privacy concerns in MPIRSResearch cognitive process of users' privacy concerns in MPIRS,including the composition of privacy concerns;influence elements of privacy concerns,and how to measure the relationship between these elements.This paper mainly studies the theory of privacy concerns and the theory of rational behavior,and based on the above theory to build MPIRS adoption behavior theory model oriented to user privacy concerns.First of all,paper summarizes six influencing factors of privacy concerns from the perspective of the user to users' privacy tendency,users' internal control,openness,extraversion,agreeableness and social groups influence.After then,paper develops the scale of independent variables and dependent variable,and defines these variables.Then,it establishs hypothesis between six factors and four-dimension privacy concerns.On this basis,paper researches the influence of each factors on the privacy concerns,and deeply analyzes user's intensity of privacy concerns,psychological of privacy preference and network behavior.Thirdly,based on influence factors analysis of users' privacy concerns,paper puts forward the model of influence factors of privacy concerns based on SEM.SEM path model verifies the hypothesis proposed and determines the structure of privacy concerns and influence mechanism.It lays the theory foundation for research of user preference model and personalized recommendation algorithm in chapter 4 and chapter 5.2.Mobile context recommended method based on the analysis of the emotional tendency under privacy concernPaper proposes a mobile context recommended methods(PS-HCF)based on the analysis of the emotional tendency under privacy concern.Firstly,paper proproses a sentiment tendency analysis algorithm based on sentiment vocabulary ontology(STAS)to improve recommendation system for predicting potential unknown user preferences.It means to use text emotional analysis technology for emotional feature extracting and to text emotional preference mining.Then,paper analyzes the influence that chapter 3 puts forward six factors of privacy concerns for mobile users adopting MPIRS.After that,paper introduces the concept of "privacy concerns intensity",and puts forward collaborative filtering method based on user combining with "privacy concerns intensity"(PI-UCF).PI-UCF uses "privacy concerns intensity" to find neighbors set,and use the known rating to predict the score of the target users.Thirdly,paper puts forward the collaborative filtering method based on user combining with context and the emotional information CS-UCF.It makes the"user-goods/services" clustering by using context similarity calculation method,which makes each subclass of "user-goods/services" have similar context.Then,a novel collaborative filtering recommendation algorithm is proposed combining the user's emotional score and the context similarity.In the end,the PS-HCF method combining prediction score of PI-UCF with prediction score of CS-UCF generates hybrid recommendation results.PS-HCF method considers "privacy concerns intensity",context and sentiment tendency information,which soloves problems such as data sparse in mobile recommendation and reduce the degree of users' privacy concerns.3.Mobile social network recommended method based on personality traits and intensity of user relationship under privacy concernPaper proposes a mobile social network recommended method(PC-MSPR)based on personality traits and intensity of user relationship under privacy concern.First of all,PC-MSPR focus on influence of three personality traits(openness,extraversion,and agreeableness)on mobile user network behavior,and integrates "privacy preference degree" into calculation model of individual personality traits.Then the above four factors are quantified,we design a calculation method of the personality traits integrating with privacy preference degree(PP-PTM).It establishes a "big-five" personality prediction model of objective network behavior characteristics and personality traits,which overcome traditional problems of mental self-test scale,such as difficult to obtain data,little data and data are unavailable because the investigation object is not serious and dishonest.Paper considers the preference degree of privacy concerns in order to obtain valuable user privacy preference.Secondly,the paper puts forward a user relationship strength estimation method(AI-URS)combining activity field classification with indirect relation.AI-URS divides the interactive activities according to activity fields and to calculate the intensity of user relationship in same activity fields.At the same time,it calculates users' unidirectional composite relationship strength according to interactive activities document in the same field.Composite relationship includes direct and indirect relations that can improve the accuracy of the calculation results,which overcomes the limitations of using direct relations.In the end,paper integrates the personality traits and user social relationships into calculation of user similarity of collaborative filtering recommendation,which soloves problems such as data sparse in mobile recommendation and reduces the degree of users' privacy concerns.4.Application on mobile personalized information recommendation service oriented user private concern problemIt applies the method and model proposed in this paper to MPIRS considering users' privacy concerns.The paper designs system framework of mobile recommendation system(MRecommend),including the network data collection module,the calculation module of emotion tendentiousness,mobile personalized recommendation algorithm module,etc.Paper constructs domain ontology based on social tagging,user context based on domain ontology,mobilephone domain ontology integrating into the comment feature,which implement multivariate information expression and management based on domain ontology of MPIRS.Finally,paper proposes privacy concerns and recommended strategy management enlightenment for mobile commerce enterprise,including the improvement suggestions in privacy mechanism,and measures and suggestions for improving the quality of mobile personalized recommendation and user adoption will of MPIRS.