Design And Implemetion Attribute Certification System Based On Data Mining And Web Trust

At present,social network service such as Facebook,Twitter has become one of the most important applications in the Internet with a rapid increase of users,which become much essential for information acquisition and social.Because micro-blog information dissemination has strong autonomy and personal selectivity,users can publish their own micro-blog content according to their own interests and hobbies and choose whether to focus on others according to the type and quality of other people's publishing content.SNS provides a variety of social networking,but on the other hand,false information on the social networking site has become a potential security risk.Users can modify their own attribute information that cause the property information is not credible.In social networks,the users attribute information crisis is becoming more and more serious.Attribute authority system is now using CA authentication method which issue a certificate for each user.This is only suitable for small closed systems.This situation is not applicable for large social networks.Based on the above background,the thesis analyzes the users'behavior based on the largest micro-blog platform and the research theory and achievements in many fields are investigated.The thesis puts forward the combination of social network user attribute information mining and PGP to approve users profiles based on WEB trust model.In the design of user attribute information mining model,the thesis takes full account of the user's attributes relevance and introduces the multi-label classification to predict the users' attributes,optimize the time performance of data mining,and improve the scalability of data mining model.In addition,by referencing PGP web of trust to verify the validity of the public key,the web trust model is proposed to provide the user with attributes authentication,which provides a safe profile authentication method.