Selecting Task-Relevant Sources for Just-in-Time Retrieval

\Just-in-time" information systems monitor their users' tasks, anticipate task-based information needs, and proactively provide their users with relevant information. The e ectiveness of such systems depends both on their capability to track user tasks and on their ability to retrieve information that satis es taskbased needs. The Watson system (Budzik et al. 1998) provides a framework for monitoring user tasks and identifying relevant content areas, and uses this information to generate focused queries for general-purpose search engines. The proliferation of specialized search engines and information repositories on the Web provides a rich source of focused information, but exploiting them depends on having methods for selecting the best sources to satisfy the user's needs. This paper describes research on augmenting Watson with the capability for automatic information source selection. It presents two automatic methods for source selection that have been integrated into Watson, compares their performance, and points to general problems and opportunities for task-relevant source selection.