Google Knowledge Vault

New Scientist reports that Google works on supplementing Knowledge Graph with a new database called Knowledge Vault. While Knowledge Graph uses information from a list of trusted sources, Knowledge Vault gathers data from the entire Web.

"Google started building the Vault by using an algorithm to automatically pull in information from all over the web, using machine learning to turn the raw data into usable pieces of knowledge. Knowledge Vault has pulled in 1.6 billion facts to date. Of these, 271 million are rated as 'confident facts', to which Google's model ascribes a more than 90 per cent chance of being true. It does this by cross-referencing new facts with what it already knows. Knowledge Vault offers Google fast, automatic expansion of its knowledge – and it's only going to get bigger," informs New Scientist.

Google already provides some answers extracted from web pages, but it's not clear if this feature has anything to do with Knowledge Vault.

It's interesting to note that the Google blog post that announced Knowledge Graph mentioned that "Knowledge Graph isn't just rooted in public sources such as Freebase, Wikipedia and the CIA World Factbook. It's also augmented at a much larger scale — because we're focused on comprehensive breadth and depth."

Google's goal has always been to use the power of the Web to extract facts and answers. 10 years ago, Google Q&A provided answers to common questions like "what is the population of Japan?" using information from the Web. Knowledge Graph allowed Google to organize information and create connections between entities and now Knowledge Vault comes full circle.

"Knowledge Vault: A Web-Scale Approach to Probabilistic Knowledge Fusion" is authored by Xin Luna Dong, Evgeniy Gabrilovich, Geremy Heitz, Wilko Horn, Ni Lao, Kevin Murphy, Thomas Strohmann, Shaohua Sun, Wei Zhang - all of them from Google. The paper explains that "Knowledge Vault is different from previous works on automatic knowledge base construction as it combines noisy extractions from the Web together with prior knowledge, which is derived from existing knowledge bases". Google's paper uses the open source Freebase graph, but it's obvious that Freebase would be replaced with Knowledge Graph in Google Search.

{ via Search Engine Land }