Research And Application Of Situational Analysis And Intention Discovery Mechanism In Multimedia Social Networks

With the rapid development of multimedia social networks(MSNs),the "explosive" growth of users and digital contents facilitates users access to digital contents,and meanwhile increases the interaction between users and users,and users and systems.How to provide users with timely and efficient personalized service in the complex interaction has become a challenge of multimedia social networking research.Therefore,the research of this subject finds the frequent behavior patterns of users in multimedia social networks based on situation analytics theory and designs a situation-based multimedia content recommendation algorithm on the basis of user's behavior model.It is convenient for user to the potential interests in the massive data.The main research contents and innovations are as follows:1.Based on situation analytics theory,a method of finding frequent behavior patterns of users in multimedia social networks is established.In the multimedia social networks,a large number of users may have different roles in different groups,and with the role in the group may cause the different intention.Firstly,based on the attributes of social networks,the Social Situ framework is established on the basis of Situ theory of Professor Carl K.Chang,and the elements in the framework are defined.Secondly,a method for finding user behavior pattern in multimedia social networks is designed based on the SocialSitu framework and the improved GSP(Generalized Sequential Pattern).Finally,by analyzing the user's history SocialSitu(t),we can get the behavior pattern of the user in the intention,according to the behavior pattern of the intention we can predict the intentions of the user's current behavior sequence.2.Design the multimedia content recommendation algorithm in the multimedia social network based on the user's behavior pattern.The current existing recommendation system does not know the user's preference for the current accessing content before the user scores or performs other operations on the accessing content,and the user's preference may change with the user's environment and the user's identity at any time.Usually in the multimedia social network,the user has their own scoring habits,or the user's score may be very casual.Therefore,it is not accurate for the traditional recommendation algorithm to predict the target user's score only using similar users' scores to the content,but the user's operations on the multimedia contents can be a true response to the user's preferences.Therefore,in order to giveusers a timely recommendation,the study proposes a multimedia content recommendation algorithm based on the user situation-aware.We can analyse the target user's preference according to the historical operations and scoring records of accessing contents of the target user's neighbor sets,and the intention of accessing content of the target user.Thus we can recommend the contents to the target user that he/she may be interested.Through experimental analysis,the recommended algorithm in the accuracy and recall rate is highly improved.Providing users with personalized service in the multimedia social networks has become the current trend of the service system.The subject mines users' behavior patterns according to users' historical behavior data and analyze the user potential preferences according to these patterns in the multimedia social networks.The main contributions of this paper are as follows:(1)extend and enrich the Situ theory;(2)derive the behavior sequence pattern of the different intention by serialization algorithm and predict the current intention of the user based on the user's behavior sequence pattern;(3)design and implement the multimedia content recommendation algorithm,which can timely perceive the potential preferences of users and provide users with timely personalized recommendation.