Research On Microblog Social Circles Mining Based On Artificial Immune Network

Various online social networking platforms in the WEB2.0 era are beginning to get hotter and hotter.More and more users choose to share information,focus on interaction,and interact with information,etc.on Weibo,and generate a lot of social information and communication information,and generate a large amount of social and communication information.If we can analyze the dissemination content and social information generated by users during various social activities on Weibo,mining the social circle information corresponding to users from Weibo's huge virtual online social network is not only beneficial to the evolution analysis of online public opinions,but also plays an important role in the field of Weibo network structure analysis and user personalized recommendation.After summarizing the research status at home and abroad,this paper believes that there are three difficult problems in the research work of Weibo social circle mining,such as Weibo data acquisition,user relationship intensity description and social circle mining algorithm design.Immune network is the main research direction in artificial immune system.Because of its adaptability and high efficiency,it has been widely used in data clustering and anomaly detection.This paper takes the user of Sina Weibo platform as the detection object,and proposes a user similarity calculation method that integrates the user's social information and interest,realizes the description of the relationship between users,and draws on the users with strong artificial immune network to aggregate to form a social circle.Firstly,in order to solve the problem of data source,this article takes a look at the use of the Web Scraper tool to obtain Weibo data with a variety of custom and complex rules.This method can effectively collect data of various web pages,and has the advantages of high crawling efficiency and perfect data type,which can satisfy the experiment.Secondly,through the analysis of Sina Weibo user relationship,the user's blog content is selected as the user's interest information,the user's attention information and fan information are used as the user's social information,and the user similarity calculation is performed to integrate the user's social information and interest similarity.To realize a description of the relationship between users,and establish a user similarity network with the user as the node,the user relationship as the edge,and the relationship strength as the weight.Then based on the idea of artificial immune network,we aggregate users with strong relationship in the user similarity network to form a social circle.This paper uses the real user data of Sina Weibo to compare the method with other social circle mining methods.The results show that the social circle mining method has better effect on the cohesion,tightness and practicability.