An Approach To Mining Influential Persons And Their Corresponding Relationships In Social Network

More and more users begin to participate in the network for information sharingand exchange with the rapid development of network technology, the concept ofsocial networks also has been presented. Users share and exchange their knowledgethat in all aspects of insight in the form of recording blogs which has facilitate thesharing and exchange of information greatly. The number of blogs has expanded asmore and more social network users appear. To be able to find the most popular usersand blogs to give other users the most valuable reference, this paper presents atheoretical perspective that mining the most influential users and their relationships inthe social network. Also, the system design and implementation is introduced in thepaper.In order to mine the most influential users and their relationships in the socialnetwork, we firstly collect data using a web-crawler, and then we analyze thecollected data to find the most influential users and their relationships. Furthermore,we provide a web-based query tool for users to show results. The first research projectis to develop a spider to collect web data on the World Wide Web, and design a dataanalyzer to analyze and extract critical information. Then users would query userrelationship in the given web page.The influence of one person is calculated by recommendation, comment and visitcount. The paper mainly introduces the algorithm design of finding the mostinfluential person and corresponding relationships. Moreover, it implements a tool forthe visualization of the results.Firstly, it describes the research background and significance of the developmentfirstly. The algorithm of MD5and TFIDF is also introduced in the paper. It describesLinux network programming, WebService and other key technologies such asSilverLight that is used in the paper, too.Secondly, carried out research on the needs analysis and algorithm design, and describes the detailed design and implementation.Finally, the paper summarizes the work we have done and follow-up to theprospect.