USER-CENTERED INDEXING FOR ADAPTIVE INFORMATION ACCESS

We are focusing on information access tasks characterized by large vol ume of hypermedia connected technical documents, a need for rapid and effective access to familiar information, and long-term interaction wi th evolving information. The problem for technical users is to build a nd maintain a personalized task-oriented model of the information to q uickly access relevant information. We propose a solution which provid es user-centered adaptive information retrieval and navigation. This s olution supports users in customizing information access over time. It is complementary to information discovery methods which provide acces s to new information, since it lets users customize future access to p reviously found information. It relies on a technique, called Adaptive Relevance Network, which creates and maintains a complex indexing str ucture to represent personal user's information access maps organized by concepts. This technique is integrated within the Adaptive HyperMan system, which helps NASA Space Shuttle flight controllers organize an d access large amount of information. It allows users to select and ma rk any part of a document as interesting, and to index that part with user-defined concepts. Users can then do subsequent retrieval of marke d portions of documents. This functionality allows users to define and access personal collections of information, which are dynamically com puted. The system also supports collaborative review by letting users share group access maps. The adaptive relevance network provides long- term adaptation based both on usage and on explicit user input. The in dexing structure is dynamic and evolves over time. Learning and genera lization support flexible retrieval of information under similar conce pts. The network is geared towards more recent information access, and automatically manages its size in order to maintain rapid access when scaling up to large hypermedia space. We present results of simulated learning experiments.