期刊名称：Bulletin of the Technical Committee on Data Engineering
出版年度：2016
卷号：39
期号：2
页码：106
出版社：IEEE Computer Society
摘要：The quality of web sources has been traditionally evaluated using exogenous signals such as the hyper-link structure of the graph. We propose a new approach that relies on endogenous signals such as thecorrectness of factual information provided by the source: a source that has few false facts is consid-ered to be trustworthy. The facts are automatically extracted from each source by information extractionmethods commonly used to construct knowledge bases. We propose a way to distinguish errors made inthe extraction process from factual errors in the web source per se, by using joint inference in a novelmulti-layer probabilistic model. We call the trustworthiness score we computed Knowledge-Based Trust(KBT). We apply our method to a database of 2.8B facts extracted from the web, and thereby estimatethe trustworthiness of 119M webpages. Manual evaluation of a subset of the results confirms the effec-tiveness of the method.
