more about topix

Filed in archive Entrepreneurship on March 10, 2004

Gary has some good inside from my favorite news aggregator Topix.

"Q. Can you share some background about how Topix.Net builds a page? Are pages built automatically or is there some human intervention? Is the technology your own? How long did it take to get it up and running?

We developed separate software modules to crawl, cluster and categorize articles. The heart of our system is a proprietary AI categorizer that uses a massive Knowledge Base (KB) to determine the geographic location and subject categorization for each story.

The final step is the Robo-Editor, which picks the best stories for display. For example, our 2004 Presidential Election page may have seen 1,000 articles for the past hour. The Robo-Editor's job is to pick the 10 best articles to show the user to give them a good overview of the news.

Our system is fully automated, there is no human involvement at any stage. We developed the technology in-house over the past two years. The AI was particularly tricky to get right, since an accuracy rate in excess of 99% was necessary to make the system useful."