Joshua Rabinowitz, Nathalie Mathe and James R. Chen
We are interested in facilitating information access from large volume of reference information contained in technical and operational manuals. When a large volume of information is used on an everyday basis to perform one’s job, the problem for users is to build and maintain an accurate cognitive model of the information in order to access it quickly. The problem is not to perform a search to discover new information, since they have already learned the information during their training. We have developed an adaptive hypertext system to help Space Shuttle flight controllers access operations documents in mission control center. We describe this intelligent system, called Adaptive HyperMan, which lets users incorporate their representation of the content and organization of documents over time. It provides sophisticated annotations and hyperlinking capabilities to end-users, and integrates an adaptive indexing and retrieval engine for managing annotations. This novel feature lets users assign topics to annotations, retrieve annotations by topics, and provide relevance feedback over time. Besides memorizing user inputs, the indexing engine also learns to generalize user inputs in order to facilitate retrieval from similar topics. We describe the Adaptive HyperMan system, then show how it provides a virtual "goody book" facility to flight controllers, and supports collaborative work.