Announcing Graph search, the young CEO said, “We’re taking a tip back to our roots”, which he said revolved around “using the social graph to make new connections.”

Describing the new search feature, Zuckerberg said that the graph consists of a billion people their 240 billion photos and over a trillion connections. All these are indexed such that they will be available for public search while ensuring that the search feature is privacy-aware.

Development of the Graph search has taken huge amount of engineering efforts and Zuckerberg added, “About 10 percent of our computing is spent on privacy checks.” Detailing the difference between Graph search and the general web search as provided by Google, Bing Zuck added, “In general, web search is designed to … return links that may have answers to the questions that you’re trying to ask. Graph search is designed to return the answer, not links that might get you to the answer.”

Mark said that the third pillar of Facebook is not just a simple search feature but, is a smarter machine with a smart search engine with capabilities to understand natural language.

“We came up with something we think is a lot more natural,” he said.

So how does this feature work? A user would type in a search query and he / she will see a drop-down menu that would show categories that the graph search understands i.e. people, places, pictures, etc. Following this the user can start searching using natural language questions such as “Who are my friends in Portland?” or “Museums in Washington, DC, my friends have been to.”

The results to these search queries are based on signals such as comments and likes as well as general content.