A Google research team is adapting Google’s pagerank to measure the trustworthiness of a page, rather than its reputation across the web. Instead of counting incoming links, the system – which is not yet live – counts the number of incorrect facts within a page. “A source that has few false facts is considered to be trustworthy,” says the team. The score they compute for each page is its Knowledge-Based Trust score.
This appears to be a direction that Google is developing. However, it appears that more work is needed before it could be scaled up and implemented in the live search engine results.
The quality of web sources has been traditionally evaluated using exogenous signals such as the hyperlink structure of the graph. We propose a new approach that relies on endogenous signals, namely, the correctness of factual information provided by the source. A source that has few false facts is considered to be trustworthy. The facts are automatically extracted from each source by information extraction methods commonly used to construct knowledge bases. We propose a way to distinguish errors made in the extraction process from factual errors in the web source per se, by using joint inference in a novel multi-layer probabilistic model. We call the trustworthiness score we computed Knowledge-Based Trust (KBT). On synthetic data, we show that our method can reliably compute the true trustworthiness levels of the sources. We then apply it to a database of 2.8B facts extracted from the web, and thereby estimate the trustworthiness of 119M webpages. Manual evaluation of a subset of the results confirms the effectiveness of the method.
SOURCES – New Scientist, Arxiv