Yahoo! Research Labs Releases Yahoo! Mindset

Sometimes at Yahoo! Research Labs, an idea is just too exciting to keep to ourselves. Mindset is just such an idea. Today we’ve posted it on Yahoo! Next for you to try out and let us know what you think.

What is Mindset? A new twist on search that uses machine learning technology to give you a choice: View Yahoo! Search results sorted according to whether they are more commercial or more informational (i.e., from academic, non-commercial, or research-oriented sources).

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.

Through the use of machine learning for text classification, we try to classify each web page in the top 100 search results for a query. Then we sort those results according to the preference you set. Mindset uses an intuitive slider in the interface, so you can set the bias for commercial vs. non-commercial results. Often, we come across a web page that hasn’t been classified yet. In those cases, Mindset tries to classify that web page in the background, so it’ll be classified along with the rest of the results next time you do the same query.

A full description of Mindset, what it is, and how it works, can be found on the Yahoo! Research Labs Mindset pages.

When using Mindset, please bear in mind it’s not a finished product. We think of it more as a technology demo and an ongoing research project. We want to determine whether: a) the underlying technology is effective and b) you like it. That’s why we’re very keen to know what you think and invite you to discuss it on our Mindset forums.

Try Mindset and then share a piece of your mind with us.

Bernard Mangold

Sr. Director, Yahoo! Research Labs

  • 38 Comments
  • Subscribe
back to yahoo! search

subscription options

Facebook Fans

latest posts

archives