Safety Threatens and Resolving Methods of Social Networks

Recently, social networks, as the most rapidly developing internet application model, is one of the best ways to communicate with friends and to share information with each others. Meanwhile, the character of close contact with the real world and open nature of platform makes the social networks easy to be attack. After analysis of the social networks, we found factors which result in SNS security vulnerability as follows: wide connection of Internet users, unlimited spread of credit, open feature of SNS platform, inadequate safety precautions, diversity ways of communication and huge network traffic. Fake (stolen) account with high credit is the necessary condition of an attack, also the key factor of vulnerability. Therefore, we propose a credit-based account security detection framework, which merge intrusion detection system and text filtering system into this framework to enhance the ability of detecting malicious account, so as to effectively prevent phishing, Trojans loading, zombie attacks and other threats.