Home People UW Professor Oren Etzioni, inventor of Metacrawler, NetBot, and now Hamlet, is...

UW Professor Oren Etzioni, inventor of Metacrawler, NetBot, and now Hamlet, is a Virtual PhD of Webology

“To buy or not to buy, that is the e-commerce question.” It is a most ominous and confounding question among Internet shoppers when attempting to fathom the depths of airline ticket pricing. Depending on the time of day you book a flight online, the price of a seat can fluctuate wildly, changing by hundreds, if not thousands, of dollars in a matter of hours. For example, the price for a recent round trip airline ticket from Seattle to Washington, D.C., soared as high as $1,668 compared to a temporary low of $281, all in the same 24-hour period.

But what if you could crack the algorithms that the airlines use to set ticket pricing in order to understand the best time to buy? By studying past trends in pricing variation and establishing the resulting patterns, the formulas could then be used to predict future price fluctuations.

Seeking to package this combination of logic and predictability in the otherwise unwieldy domain of “dynamic pricing” is the brainchild of University of Washington professor, Oren Etzioni. It has led to the creation of Hamlet, Inc., the latest company in a series of info-vative start-ups from the UW computer scientist who could be dubbed Seattle’s honorary Professor of Webology. Hamlet, Inc. has just named a president, Mike Riccio, has a round of venture funding behind it, and is taxiing for a takeoff on a not too distant runway. Indeed, you could soon be thanking Hamlet, (and Etzioni) for making your travel plans a lot more affordable, if all goes according to plan on your next booking.

Along with his academic post at UW and his consultative role as a Venture Partner for the Madrona Venture Group, Etzioni was also the Chief Technology Officer and board member of Go2net, Inc. (which was acquired by Infospace in 2000), and a co-founder of Netbot, Inc. At Netbot, Etzioni helped to conceive and design the web’s first major comparison-shopping agent (a price “bot” that foreshadowed MySimon and which was acquired by Excite in 1997).

In 1995, Dr. Etzioni and his student Erik Selberg developed MetaCrawler, still the web’s premier Meta-search engine – now being run by Infospace.  Oren is also a co-founder of Clearforest, an Israeli startup, which is now an international leader in text mining.

His latest initiative is the “Semantic Web,” so-called because it is based on a more orderly descriptive language that creates a more transactional “table of contents” for online information, arranged by a glossary created by Web users worldwide.

We caught up with Oren to see how he manages to get it all done.

Seattle24x7: You’ve played a pioneering role in the development of the Internet with your concentration on search algorithms, and comparitive pricing. What is the relationship for you between Academia and the birth of new commercial concepts on the Web?
Etzioni: We are very proud of the fact that we were the first people to do metasearch and comparison shopping in a real way. NetBot was the first to do any kind of serious comparison shopping. My Simon, which is owned by CNET, and DealTime are direct descendants, intellectually speaking.

Our academic research doesn’t really track with the hot commercial trends though. Our research is purely curiosity-driven, tracking what we think might make a difference twenty years from now. There’s not a lot of correlation with what’s hot in the market today, most of which was hot in academia a while ago. For instance, the fundamental ideas on search engines are based on information retrieval research that was done in the 60’s and 70’s. Google’s innovations today are based on algorithms that were defined in the 80’s and 90’s.

Seattle24x7: How do you manage to coordinate academic life with your business creation and development activities?
Etzioni: I am full-time at the university. These other projects are my secondary commitment. We are allowed to spend a day a week consulting, so I squeeze them in each week. My primary focus is academic, the current interest outside of my university work is Hamlet.

Seattle24x7: Can you fill us in on what’s happening with Hamlet?
Etzioni: Hamlet was a technology that was developed at the University of Washington Computer Science Department. We founded the company called Hamlet, Inc. which licensed the technology and hired a CEO, Mike Riccio. We have some initial funding from Madrona Venture Group. Our focus is the travel industry at this point. [Not only airline tickets but also] hotels, vacation packages, car rentals, there’s a lot of places where price prediction would be of interest.

Seattle24x7: Do you see an application for predictive pricing in other markets?
Etzioni: People asked us couldn’t you just use this for the stock market, and the answered is “no.” The behavior of airline prices, and the prices of goods in the travel industry, are set by algorithms. In the case of the stock market, it’s based on people buying and selling. So it is very different, The travel industry is a $550 billion dollar market. I think we’ll be busy for a while.

Seattle24x7: You’re also doing a lot of work on “The Semantic Web” and “Semantic email. ” What’s that all about?
Etzioni: Here’s a quick example. Suppose you want to schedule a meeting with some people. If you’re all using Outlook than Outlook provides a few services to do that but if you’re all in different places and different time zones and trying to set up a telephone conference, how do you do that?

We’ve provided a simple service that runs on our servers at the university. Anyone can use it. There are all kinds of processes: voting, fist come first serve, giveaways, meeting coordinations, these are all available. These processes are all what I consider to be on-ramps to the Semantic Web. You can learn more at http://www.cs.washington.edu/research/semweb/email.html [24×7]

Oren Etzioni has helped the UW become a spawning ground for breakthrough ideas in Web search like MetaCrawler and Netbot, and now, predictive pricing with Hamlet.