Google Launches New AI ‘RankBrain’ to Answer Ambiguous Queries

A recently deployed artificial intelligence (AI) system is now helping to power a significant portion of Google searches.

Of the millions of queries it receives each second, about 15 percent of them are completely new to Google.

But what happens when you ask Google something it’s never seen before?

This announcement follows many developments from Google in the artificial intelligence space, which includes the acquisition of the AI company DeepMind Technologies in 2014. The AI links the written language that users send to Google with computer-friendly mathematical expressions and models that the search engine’s servers can actually understand.

Bloomberg, reporting on the new AI, writes that a “very large fraction” of searches in the past few months have been answered by the system.

Corrado says RankBrain is different from the “hundreds” of signals and technologies that contribute to Google’s Search algorithms in that it actually learns and improves over time. The AI system known as RankBrain is being used to generate relevant search results for unique queries, according to reports.

For example RankBrain can better handle questions like “What’s the title of the consumer at the highest level of a food chain?”

Corrado said that has quickly become the third most important signal, especially when it comes to processing ambiguous queries. It also guesses which words and phrases are similar in meaning to each other so it can help simplify queries for Google to look up.

Corrado notes, “Machine learning isn’t just a magic syrup that you pour onto a problem and it makes it better”. And now, RankBrain appears essential to Google.

But unlike other Google algorithm signals, like Panda, Penguin and Hummingbird which are ranking signals to find quality websites and eliminate spammers, RankBrain’s job is to process never-before-seen and more complex questions coming from Google users. “It took a lot of thought and care in order to build something that we really thought was worth doing”. The humans only managed 70 percent.