A Maximum Entropy Weighted Fact-Finder in Information Network

Abstract

In information network, different sources publish facts with different degrees of credibility and originality. To predict the truth values of the facts, several fact-finder algorithms are suggested which iteratively compute the trustworthiness of an information source and the accuracy of the facts it provides. However they ignore a great deal of relevant background and contextual information. In this paper, we propose a novel maximum entropy weighted method to processing trust analysis, which allows us to elegantly incorporate knowledge such as the attributes of the objects and the implications of the sources. Experiments demonstrate that our algorithm significantly improves performance over existing.