Issue No. 05 - September-October (1997 vol. 12)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/64.621224
InfoFinder is an intelligent agent that learns user information interests from sets of messages or other online documents that users have classified. While this problem has been addressed by a number of recent research initiatives, InfoFinder's approach is innovative in a number of ways. First, the agent uses heuristics to extract significant phrases from documents for learning rather than using statistical techniques. This enables it to learn highly general search criteria based on a small number of sample documents. Second, the agent's induction algorithm is based on the observation that sample documents in such an application will not be uniformly distributed, because of the fact that users will tend to classify positive examples while browsing while classifying negative examples only when the agent makes a bad recommendation. Third, the agent learns standard decision trees for each user category. These decision trees are easily transformed into search query strings for standard search systems rather than requiring specialized search engines, and are significantly more expressive than other representations such as positive and negative word lists.
B. Krulwich and C. Burkey, "The InfoFinder Agent: Learning User Interests through Heuristic Phrase Extraction," in IEEE Intelligent Systems, vol. 12, no. , pp. 22-27, 1997.