How do you find an answer in a sea of data? What would be the most painless way? You’d go ask someone who might know, right?
But now we’re in the Information Age. We don’t have to suffer and sift through card catalogs at the local library. Or buy stacks of encyclopedias from a creepy salesman. We have Google. You can simply type your question into Google, and it will spit out a litany of possible results in in some absurdly short interval of time. It’s fast and accurate.
Sometimes.
How does Google know you are looking for information on sequoia trees, instead of the Toyota Sequoia? How can it understand common sense queries? How are we going to build the search tools of the future?
The key, some believe, is user modeling. The development of an intelligent agent that observes what your interests are and the way you think, and estimates what you would want to search for. To borrow a term from Frank Herbert’s Dune, a mentat.
In a way, Google already compensates for its artificial intelligence by harnessing the power of human minds. Google merely observes how us humans build our websites and link to others. Using this kind of citation analysis, Google knows what are the most popular sites.
Techno-profiling is still very crude. It is still doesn’t understand natural language. It needs to learn some common sense. “The Web needs to go through the infantile process of self-discovery. The Web doesn’t really understand itself. There’s lots of information on the Web, but not much “information about information,” also known as “metadata.””
Paul Allen’s Vulcan Inc. are tackling the same issue with a research tool called Project Halo. A “Digital Aristotle”, filled with scientific information, that can respond to natural language queries. Unfortunately, like that guy you used to know in university with a photographic memory, it’s doing well when asked to simply regurgitate data from its memory banks, still having problems understanding application-based questions.
It also suffers from the age old problem of content quality. It’s first prototype failed because it wasn’t taught chemistry correctly.
Auto-trackback from memigo.com
Washington Post article found in memigo. It was referred to by 5 source(s).
Follow trackback to find related articles…