Skip to main content

Google Hones Search With “Knowledge Graph”

Google has revealed plans to tweak its search engine and make it a more "human" experience via the introduction of a "knowledge graph" system, which will roll out in the US first, and then worldwide.

The knowledge graph is all about giving its search engine a better understanding of search terms and their relationships, rather than just simply and mechanically sift through keywords.

Google gives the example of a search for the "Taj Mahal". Someone searching for that might be seeking information on the famous landmark, the musician, or their local Indian restaurant.

The search engine will be able to pick out these most common motifs for that particular search, and allow the user to narrow the search results to one of these specific elements with a single click. Hence results are easily made far more relevant, and require less work hunting through on the user's part.

The knowledge graph will also be used to highlight key topics around a search result, picking out facts based on how many users have hunted for them following a search. In other words, the most likely relevant linked content will be presented.

Google's knowledge graph draws from a number of sources, not just Wikipedia and public resources, and contains some 500 million objects at the moment, as well as 3.5 billion facts regarding the relationships between these objects.

There's still plenty more work to be done, though. Amit Singhal, SVP, Engineering, Google, commented in his blog post: "We're proud of our first baby step - the Knowledge Graph - which will enable us to make search more intelligent, moving us closer to the ‘Star Trek computer' that I've always dreamt of building."

Source: Google Blog