The California-based Internet titan is intent on adding "semantic" capabilities to automatically comprehend meanings of phrases and questions to better fetch the online information being sought.
"Right now, our understanding is pretty darn limited," Google fellow and online search veteran Amit Singhal said of online search Thursday in a message posted at the company's Google+ online social network.
"Ask us for the 10 deepest lakes in the US and we'll give you decent results based on those keywords, but not necessarily because we understand what depth is or what a lake is," he explained.
To make search smarter, Google is tapping into the virtual brain of a Freebase database of knowledge regarding what things are and how they relate to one another.
Google acquired the "open-source knowledge graph" when it bought San Francisco-based Metaweb Technologies in 2010 for an undisclosed sum.
"In the time since, we've grown (Freebase) from 12 million interconnected entities and attributes to over 200 million," Singhal said.
"Our vision for this knowledge graph is as a tool to aid the creation of more knowledge -- an endless cycle of creativity and insight."
Traditional Internet search formulas recognize words typed into query boxes and then deliver links to websites that appear relevant.
Google did not reveal a timeline for the evolution to "semantic" capabilities.
Google is perpetually tuning its search engine, and in January wove content from its social network and Picasa photo-sharing service into its search formula to serve up personalized results to online queries.
"Search, plus Your World" lets people signed into Google accounts get search results that include content approved for sharing by them or friends at Google+ or Picasa.
"Let me just say that every day, we're improving our ability to give you the best answers to your questions as quickly as possible," Singhal said.
"So stay tuned for updates on what will continue to be a long road ahead."