Design And Implementation Of The Community Relationship Mining System Based On Campus BBS

With the development and popularization of the Internet,social networks are becoming more and more important in people's life.Board System Bulletin(BBS)is one of the earliest and most typical platforms in the Internet,and it has been gradually developed into an important platform for people to express ideas.Project in this thesis mainly do the analysis on the data from the campus BBS.As an important part of the Internet and the University network,it has a strong purpose,which makes it very important.This thesis studied and implemented a system based on replies and interactive behavior between BBS users,and also analyse the text from the BBS,the system has function of data acquisition and storage,data analysis and mining,and the display platform.It can work out the association between the users from analysing the interaction and text in the BBS.It can also labeling the communities to show their characteristics and then show them on a platform.The system has an important significance for the evolution and development the public opinion monitoring,as well as for the management and construction of the school.The research work of the communitiy mining system is as follows.1.Make use of the public API of the BYR Forum to crawl for the data,and use the MongoDB to store the data.Whiling crawling the data,we need to do some encoding and checking works to make the complex data more normalized.After the research work,I decided to use the MongoDB to improve the operating efficiency.2.The system works out the final community results by analysing the interaction and text content from the BBS.On the one hand,I use the GN alforithm to work out the relationship between users and then find out the community.On the other hand,I use the text and emotional analysis tools to work on the existing community to find out their characteristic.