semweb2slide31

From CommerceNet Wiki

Jump to: navigation, search

Image:semweb2.slide.31.png

Prev Slide Index Next Slide

Even without adding semantics, AI can still improve the selectivity of RSS feeds.

Here’s a mockup of a “Learning Browser” being developed by Dan Gould and Brian Norwood, with support from CommerceNet Labs. Unlike traditional aggregators that summarize new feeds by listing headlines or subject lines, Dan and Brian present the user with a personalized ‘zine. In the spirit of the Media Lab’s “Daily Me”, a ‘zine is a digest of the latest feeds relevant to the task at hand – whether it’s researching a report, following a sport, or in the example shown, planning my trip to Pittsburgh. Over time, the browser refines its model of my interests and modifies what’s presented, based on implicit feedback such as what I click on and what I don’t, how long I spend on an article, what I recommend to my friends, what they recommend to me, as well as what’s happening in the world. We believe most sites can benefit from this kind of implicit feedback, from search engines to dating services.

Retrieved from "http://wiki.commerce.net/wiki/semweb2slide31"
Personal tools