A Method for Extracting Influential People for the Improvement of Contents

摘要

It is very important to find influential users who have useful information as well as extract helpful knowledge that affect the efficiency of user's intellectual work; however, it is difficult to evaluate how such helpful comments and conversations from influential users can affect works of other users. In this study, we propose a method for extracting user comments that influenced other users to improve their contents such as a presentation slide by analyzing correlation between user's comments and improved parts of contents. Our method allows us to make actual evaluation to influential users for some specific users. In the experiment, we evaluate the feasibility of the proposed method using the actual communication history that is collected from a communication service, Slack.