Current Attitude Prediction Model Based on Game Theory

摘要

Social interactions on online communities involve both positive and negative relationships: people give feedbacks to indicate friendship, support, or approval; but they also express disagreement or distrust of the opinions of others. One's current attitude to the other user in online communities will be affected by many factors, such as the pre-existing viewpoints towards given topics, his/her recent interactions with others and his/her prevailing mood. In this paper, we develop a game theory based method to analyze the interactive patterns in online communities, which is the first in its kind. The performance of this prediction model has been evaluated by a real-world large-scale comment dataset, and the accuracy reaches 82%.