Design And Implement Of Friends Inference Model Based On Trust For Mobile Environment

Different from the traditional Internet, users in mobile communication network correspond with their mobile phone numbers, thus the real-time user information is more accurate. For instance, we can get user’s accurate location information and real-time communication record via mobile communication network. User’s communication behavior is one of the most direct data sources of user’s social relationship, which represents his social activities. So we can dig out the social relationship between users by analyzing their communication behaviors. Thus this ; roiecthas proposed that we should do research on "social networks based on mobile users’behavior" in depth, so as to design and realize a’’mobile social network system’". The mobile social network system is mainly based on historical records of mobile users’interactions and location information, and then we can dig out the friends’relationship by analyzing influence and trust degree between mobile users, so as to provide a reference to personalized service and applications in mobile network for precise recommendation or retrieval.This paper has built a mobile social network based on a trust degree model in mobile communication area, which is established by analyzing the communication behavior and location information of mobile users, including analyzing communication behavior and location information of mobile users, designing trust degree algorithm such as direct trust degree algorithm, indirect trust degree algorithm and trust degree fusion algorithm; designing adaptive trust degree algorithm by analyzing the dynamic characteristics of trust degree; testing these algorithm via using open data sets. The experiments show the rationality and practicability for the proposed trust calculation model. In this paper, the first chapter briefly describes the background, research meaning of the friends relationship inference pattern based on trust degree in mobile environment as well as the main content of this paper. Second chapter has made some brief description to the basic concepts and data sets involved in the system design. Chapter three describes the demand analysis of each functional module and the system development environment. Chapter four describes the location of our system in the prototype system-recommender system, the system module functions as well as its data processing flow. Chapter five describes the design and implementation of each module in detail, especially the solution for direct trust degree, indirect trust degree, trust degree fusion and self-adaptive integration. Chapter six tests and analyzes the rationality and feasibility of trust degree module. Chapter seven summarizes the work and makes some prospects for future work.