Semantic Search and the Law

March 15th, 2012

Google is moving from keyword searches to what it calls a “semantic search.”

Westlaw and Lexis are (at best, lousy) keyword search engines.

Google search will look more like “how humans understand the world,” Mr. Singhal said, noting that for many searches today, “we cross our fingers and hope there’s a Web page out there with the answer.”

When you search something on Westlaw, you simply hope that there is some opinion or treatise out there that answers your question. Anyone who has done any research knows this is seldom the case, and usually you have to dig around a lot until you find something kinda similar, and then figure it out. A semantic search would work something like this:

Under the shift, people who search for “Lake Tahoe” will see key “attributes” that the search engine knows about the lake, such as its location, altitude, average temperature or salt content. In contrast, those who search for “Lake Tahoe” today would get only links to the lake’s visitor bureau website, its dedicated page on Wikipedia.com, and a link to a relevant map.

For a more complex question such as, “What are the 10 largest lakes in California?” Google might provide the answer instead of just links to other sites.

To provide answers that aren’t already in Google’s ever-expanding database, the company will blend new semantic-search technology with its current system to better recognize the value of information on websites and figure out which ones to show in search results. It would do so by examining a Web page and identifying information about specific entities referenced on it, rather than only look for keywords.

Here are key differences:

Keyword Search

  • Determines the importance of websites based on the words it contains, links to those sites and dozens of other measures.
  • Also factors in the person searching, such as his location and the time of day.

Semantic Search

  • Refers to the process of understanding the actual meaning of words.
  • Can differentiate between words with more than one meaning, such as the car brand ‘Jaguar’ and the animal ‘jaguar.’

When I talk about a Siri for the Law (Harlan, I call it), this is what I’m talking about.