An algorithm for finding frequent patterns in social media stream

Social media has substantially changed the way organizations, communities, and individuals communicate. However, the existing studies do not mention how the knowledge hidden in the gigantic volume of collected data would be found out. Efficient mining methods for social data stream are therefore still in great demand, especially for sharing common information among different services. In this paper, we propose an algorithm composed with FP-tree and LRU structure to discover frequent rules from the social media streaming environment. In the results of the performance evaluation, the runtime of UPTree is faster than existing FP-tree based approach (average 60%) on SNS environment.