Helping users find the most knowledgeable, trusted, relevant and helpful information they seek for a specific purpose, at a specific time, is what Google and Bing or any search service is all about. For that reason, Google appears to be taking steps to test newer forms of “knowledge-based” trust factors to find the most accurate information about specific topics. According to a report last week by New Scientist, a team of research scientists at Google has published a paper (PDF) explaining the idea of Knowledge-Based Trust (KBT), an approach to determining the quality of web pages that seeks to measure how accurate they are—not merely how “popular” they may be.
According to the scientists, “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.”
Search Engine Land’s Matt McGee says Google has been building a massive database of known facts for years, and in 2012 introduced its Knowledge Graph. That’s the source of those information boxes that show on the right side of Google search results (primarily) for searches involving people, places and known entities.
The Google scientists say they’ve applied Knowledge-Based Trust testing to enough web pages to reliably predict the trustworthiness of 119 million webpages and 5.6 million websites.
But don’t look for Google to throw out its algorithms anytime soon—if ever.
According to McGee, the KBT concept may not work uniformly across the internet, since many web pages do not exist to share facts and aren’t about the types of entities that exist in a Knowledge Graph-style database. In other words, it is more likely to see some aspects of the engineers’ findings to be incorporated into Google search, but don’t look for it to replace the basics of how Google works today.
Update: A few days after the paper appeared, Google’s John Mueller told a Google hangouts audience that the paper is not something they currently use to rank web pages.