A New Model for Classifying Social Media Users According to Their Behaviors

User generated content in online social media is growing rapidly, which makes it hard to be validated and verified. Facebook and Twitter are the most popular social media that are being used as a means of social communication and sharing thoughts, knowledge and even news. Information in these social networks can be generated by anyone from anywhere in anytime. Classifying such huge information using traditional data mining classification algorithms is time consuming task which needs huge processing and memory space. In this paper, we propose a new threshold-based approach for classifying information in social network that can give accurate result similar to support vector machine (SVM) with less processing time and consuming less memory space compare to SVM. We applied our experiment on Twitter accounts by monitoring KSU, SPP_ KSU and SSS_ KSU followers' accounts and compare our results with SVM results that applied by Research Chair of Pervasive and Mobile Computing (CPMC) in KSU on the same followers' accounts.