Sheila Denn Title: Making Search Results More Useful: The Role of Categorization Abstract: As information retrieval activities have increasingly shifted from information professionals using proprietary systems like Dialog to end users using search engines on the Internet, we need to work harder to ensure that the results displayed to these end users are effective in helping them satisfy their information needs. It is well documented that users employ only a very small number of terms (between two and three) for their initial searches, and are usually faced with a simple listing of results ranked according to search algorithms that are invisible to the user. One way to alleviate this problem of cognitive overload for the user could be to use linguistic techniques like latent semantic indexing (LSI) to assign documents into browseable, pre-defined categories. This could provide a more natural way for the user to interact with search results, allowing her to navigate through a hierarchy to get at the results that are most likely to be of use, and helps to ameliorate the problem of multiple word senses by grouping items together in a meaningful way. By extension, this use of categorization to get at word sense could be a step along the path to full personalization of search functionality based on individual language usage.