Personalized Rumor Refutation Through Graph Regular Pattern

Rumors spread rapidly across online social media and may cause severe economic loss and destabilize society. Considering different people are susceptible to different rumors, this paper proposed PRRM, a personalized rumor refutation mechanism based on individual interest and background knowledge. PRRM develop a regular rumor propagation pattern to characterize rumor spreading. Then we design a reverse pattern querying technique to mine the regular pattern and a regular pattern matching algorithm to identify the people who are more susceptible to given rumors. Using real-life Sina Weibo datasets, we experimentally verify the efficiency and effectiveness of our methods.