Beyond content-based retrieval: modeling domains, users and interaction

Most research in information retrieval has focused on the problem of matching user queries to stored document representations. In this talk, I will take a step back from the core matching algorithms and look at other sources of information that can fruitfully be brought to bear to improve information access in a variety of applications. Important sources include: a) information about the world of objects (domain modeling); b) information about the user (user modeling); and c) improved interactive interfaces for information access and management.