Yahoo 'Mindset' Classifies Search Links
Yahoo has taken a new mindset on how it classifies search results. Yahoo Mindset is a text classification system that segregates Web pages into two categories: commercial or information. This is accomplished by using advanced algorithms that "learn" how to interpret unclassified Web pages by examining human behavior.
Yahoo Labs is providing a beta preview of the technology that is more or less in its infancy. From a technical standpoint, the machine learns like a baby does mimicking adult behavior. Variable training algorithms and tests sets provide what Yahoo describes as a "seed set" of basic knowledge that advances the learning process through trial and error.
When a test fails, the machine reverts back to its training set and tries again; when it does right, the knowledge in the training set grows.
Yahoo developers explained why Mindset was created in the Yahoo Search Blog. "Sometimes you want to buy stuff and sometimes you just want to do research. In a typical search page, results point to commercial pages that are mixed together with non-commercial pages, so it's harder to find the type of information you're looking for. Mindset is our attempt to help solve that problem."
The Mindset demo has a "slide" that tells the users whether search results are commercial or non-commercial. The blue bar is commercial; the orange bar is informational. The bar's position along its scale infers how sure the AI is about its classification.
More technical information may be found at the Mindset research site.
For now, Yahoo is gathering feedback to determine whether or not its users like the demo and if they find it to be useful. "This is strictly a research tool that came out of Yahoo! Research Labs in Pasadena, CA. We're constantly testing new technologies and ideas in Search and other areas to ensure that our users are getting the best experience possible," a Yahoo spokesperson told BetaNews.