This paper introduces a new approach to provide users with solutions to explore a domain via an information space. A key point in our approach is that information searching and exploring takes place in a domaindependent semantic context. A given context is described through its vocabulary organised along hierarchies that structure the information space. These hierarchies are simplified views on a more complex domain specific ontology, that form a shared and coherent background knowledge representation. So the system benefits from the combination of two innovations that make the exploration of the document space more effective. First, hierarchies (extracted from the ontology) provide with a language and synthetic representation to be explored by the users to express their information need. Additionally, a visual interface presents answers to their queries using multi-dimensional analysis and a global visualisation of document collections. At both stages, ontology is the key structure that guides a meaningful browsing for query formulation and for the document set exploration.

Abstract. Conventional document search techniques are constrained by attempting to match individual keywords or phrases to source documents. Thus, these techniques miss out documents that contain semantically similar terms, thereby achieving a relatively low degree of recall. At the same time, processing capabilities and tools for syntactic and semantic analysis of language have advanced to the point where an indextime linguistic analysis of source documents is both feasible and realistic. In this paper, we introduce document dimensions, a means of classifying or grouping terms discovered in documents. Using an enhanced version of Jakarta Lucene[1], we demonstrate that supplementing keyword analysis with some syntactic and semantic information can indeed enhance the quality of information retrieval results. 1