Ranking social news articles based on voter credibility

Abstract

Given the wide spread of web-based tools and social news media services which are facilitating grassroots journalism, there is a growing interest in selecting credible news content among a huge number of articles. Currently, most of social news services rely on reader votes to select articles for their front pages. However, the fundamental problem is that users' votes often stand for popularity rather than credibility. In this paper, we propose a system to address this problem using a weighted voting system. Specifically, we trace thousands of users and their votes, differentiating them depending on how credible the articles voted for are. We then calculate each user's voting credibility and use it as the user's voting weight in our system. The results indicate that our method performs better in selecting credible news articles than other methods replying on a "one person, one vote" system. The results suggest feasible solutions to problems in social news media concerning media credibility. Copyright © 2010, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.