On the prediction of re-tweeting activities in social networks: a report on WISE 2012 challenge

摘要: This paper reports on our participation in the Data Mining track of WISE 2012 Challenge. The challenge is to predict volume future re-tweets and possible views for 33 given original short messages (tweets). Towards this, we compare contrast four different methods highlight choice accomplishing this challenge. first method a naive approach that discovers regression function based popularity network connectivity. second build classifier learns classification model user's preferences categories topics. third focuses simulation leverages Monte Carlo simulate re-tweeting paths starting from root message. fourth uses collaborative filtering recommendation model. results these are compared terms their effectiveness efficiency. Finally, insights into predicting message spreading social networks also given.