Google mines social network data for semantic search
Google has revealed a new semantic search tool to make it easier to find information, rather than links to pages – helped by data on Google+ profiles.
The Knowledge Graph looks to understand what a queried object is, rather than treat it like a string of random characters – such as knowing the difference between the Taj Mahal monument and a local restaurant with that name.
By understanding “the nuances in their meaning, the way you do”, Google will be able to provide more intelligent search, said Amit Singhal, senior vice president of engineering, in a post on the Google blog.
“It’s why we’ve been working on an intelligent model – in geek-speak, a ‘graph’ – that understands real-world entities and their relationships to one another: things, not strings,” said Singhal.
To start, the Knowledge Graph will find its way into search results in the form of a side panel which surfaces extra information about a subject. Google showed an example using Leonardo Da Vinci, which displayed a photo, small biography, paintings he created, and other painters from the same era alongside the standard search results.
Google will be using public sources such as Freebase, Wikipedia and the CIA’s World Factbook, but will also be creating its own database – which already has 500 million objects and 3.5 billion facts about them. Google has since confirmed to PC Pro it will be also using data from Google+.
“We’re blending and combining sources, including Knowledge Graph connections and Google+ profiles, to return the most relevant content,” a spokesperson said in a statement. “Part of this is about giving Google a good understanding of people and their interconnections.”
Google has started rolling out the new tool in the US, and there was no word when it would arrive on the UK version of the site.