KDnuggets : News : 2007 : n23 : item17 < PREVIOUS | NEXT >


From: R
Subject: New York area: Chief Scientist at a well-financed and rapidly growing venture

Chief Scientist. Competitive salary, incentive compensation, and equity in a well-financed and rapidly growing venture. PhD preferred. New York area.

Necessary Skill-sets

  • Algorithm design and construction
  • Experience working with sparse and noisy data
  • Statistical modeling and data mining
  • User modeling in a dynamic, changing system
  • Taxonomy construction
Problem Area
Our problem deals with users navigating the Internet space:

1) The users: A user has a finite set of characteristics that are each binary in nature. The user pool approaches fifty million and will continue to grow. While we do not know the specific characteristic assignment of each user, we know the approximate distribution of each characteristic within the user pool at large.

2) The world: The Internet is an ever-changing and dynamic system.

3) Observable actions: We can view user behavior on the Internet through click-stream data, but we are limited to an incomplete view for specific domains only.

4) The test set: We have a test dataset containing both the characteristics of each user as well as a complete view of each userís click-stream behavior.

Question: How do we return the X users most likely to have a particular characteristic, where X must exceed 2 million?

Given the specific nature of the problem, a Chief Scientist would need direct experience in the online world.

Due to high market demand for our capability, the Chief Scientist will need to have an orientation towards implementation, not long-term theoretical research.

Contact Information
Please send CV to [email protected]

Bookmark using any bookmark manager!

KDnuggets : News : 2007 : n23 : item17 < PREVIOUS | NEXT >

Copyright © 2007 KDnuggets.   Subscribe to KDnuggets News!