The Identification Of Social Network Organization Members And Interests Mining

With the development of Web2.0 era,Social networking platforms have become an integral part of Internet services,Internet users are actively joining this new information exchange platform.The number of social network users is growing rapidly,and it also led to an explosive growth of the amount of user information in social networks.Therefore,in recent years,more and more scholars have carried out research on social networks.The identification of organization members in the research of user relations and interests mining in the research of user characteristics are a very valuable research issue in the social network field.The research on the identification of organizational members is still in its infancy.Nowadays,most of the identification methods of organizational members only examine fan relationship between users and organizations,and there is a lack of research on the diversity of user behaviors and the influence of different user behaviors on user relationship discrimination.The research on user interests mining in social networking has already had a lot of achievements.However,the existing research lacks the exploration of the inherent relationship of interests and does not apply the association to the mining of interests.The main contents of this paper are as follows:(1)A method for identifying organization members based on the behavior of social network users is proposed.According to the characteristics of social network Twitter and the behavior attributes of its users,this method defines a variety of identification factors to quantitatively describe the behavioral relationships between users and user groups and summarizes the basic rules for determining social user relationships,so that the definition of human relationships in the social network transforms into the iterative computation of the model.Then the empirical research on the recognition model based on the real data of social network is carried out to discuss the degree of influence and the optimal combination model of multiple recognition factors in the recognition process.(2)A method of interests mining based on association rules is proposed.First of all,based on the profile of Linked In users,interests are modeled,and different types of interest strings are standardized using uniform interest items,and the high frequency interest items are excavated as research objects.Based on this,a correlation analysis model of interest is established,and association rules of interest are generated based on real user data of Linked In.Finally,we use the association rules of interest to improve the original Twitter user interest mining method based on word frequency.(3)Combining the above two points,the author develops a personal attribute recognition system based on massive social network data and applies the method of mining membership identification and hobby mining to the system.The system divides social network users into organizational units and studies their personal attributes based on the excavated user interests and hobbies.The method proposed in this paper can identify users belonging to the same organization in a complex social network and accurately excavate the user's interests.The research results can be used not only for commercial activities such as advertisement delivery and friend recommendation,but also for the study of social network users' behavior characteristics and attribute characteristics.