Exploring Bibliographic Collections using Concept Hierarchies.

This paper presents a new approach for browsing and mining document collections. The study is based on a data collection of 10,000 articles extracted from the ADS, corresponding to all papers authored or co-authored by French astronomers during the period 1996 to 2000. A key point in our approach is that information searching and exploring take place in a domain-dependent semantic context. A given context is described through its vocabulary organized along different concept hierarchies that correspond to different points of view. These hierarchies structure the information space. Moreover, they provide the query language for users and allow them to explore the vocabulary of the domain before they express their information need which is helpful when the information need is not well defined.