The Lemur Toolkit (www.lemurproject.org) has become a standard resource for researchers in the field of information retrieval. The combination of very efficient indexing tools, support for a variety of retrieval models, and a powerful query language has enabled a wide variety of research projects. In this collaborative project, W. Bruce Croft, University of Massachusetts - Amherst and Jamie Callan, Carnegie Mellon University are extending the toolkit with new features, search techniques, efficiency improvements, and evaluation measures to broaden the range of research it can support and to keep up-to-date with the latest research. One feature in particular is a toolbar for the acquisition and analysis of user data, such as clickthrough, that is the basis for improving many Web search techniques. This new feature helps the academic community study these techniques and will include explicit controls for safeguarding the privacy of the data and anonymizing subsets for research. The toolkit is already in use in educational and research environments world-wide and the extensions are likely to support new research leading to more effective search engines. This in turn will have a significant impact on Web users by improving the quality of retrieved results.
