An integrated Approach for Semantics-driven Information Retrieval

This paper presents a user-centered and holistic approach in order to make large and networked information spaces semantically available and researchable by including human use and interpretation. Application areas are specialized document collections such as intranets, complex portal sites and knowledge databases. The approach combines conventional full-text search with semantic retrieval methods on the basis of a learning probabilistic ontology, which abstracts and condenses an information space. New ontological concepts are generated by observing users information filing behaviour when creating bookmarks. The various access mechanisms offered are integrated into one consistent user interface. The main feature of the user interface is the visualization of the ontology. This allows browsing on a structural level, the contextualization of full-text search results by the concepts of the ontology and supports the visual construction of semantic queries for an accurate search. By using these tools, an information need can be iteratively specified more precisely. Extensive user testing shows performance advantages of the interface and attests usefulness in the application area. With in another study the feasibility and high quality of the collaborative knowledge acquisition by observing users filling behaviour could be demonstrated.

[1]  Wolfgang Strauss,et al.  Netzspannung.org - An Internet Media Lab For Knowledge Discovery In Mixed Realities , 2001 .

[2]  Ramana Rao,et al.  Laying out and visualizing large trees using a hyperbolic space , 1994, UIST '94.

[3]  Steffen Staab,et al.  Mining Ontologies from Text , 2000, EKAW.

[4]  Steve Jones.  Graphical query specification and dynamic result previews for a digital library , 1998, UIST '98.

[5]  Marc M. Sebrechts,et al.  Visualization of search results: a comparative evaluation of text, 2D, and 3D interfaces , 1999, SIGIR '99.

[6]  Tiziana Catarci,et al.  An Ontology Based Visual Tool for Query Formulation Support , 2004, OTM Workshops.

[7]  Michael Wurst,et al.  Entdeckung und Nutzbarmachung von stillem Wissen in heterogenen Expertengemeinschaften – das Projekt AWAKE (Uncovering and Utilization of Tacit Knowledge in Heterogeneous Expert Communities) , 2003, i-com.

[8]  Jürgen Ziegler,et al.  Matrix Browser - Visualisierung und Exploration vernetzter Informationsräume , 2002, Mensch & Computer.

[9]  Richard Fikes,et al.  The Ontolingua Server: a tool for collaborative ontology construction , 1997, Int. J. Hum. Comput. Stud..

[10]  Nicola Guarino,et al.  Formal ontology, conceptual analysis and knowledge representation , 1995, Int. J. Hum. Comput. Stud..

[11]  Fabrizio Sebastiani,et al.  A Note on Logic and Information Retrieval , 1995, MIRO.

[12]  G. W. Furnas,et al.  Generalized fisheye views , 1986, CHI '86.

[13]  Jian-Yun Nie,et al.  Towards a probabilistic modal logic for semantic-based information retrieval , 1992, SIGIR '92.

[14]  Michael Uschold,et al.  Ontologies: principles, methods and applications , 1996, The Knowledge Engineering Review.

[15]  Christian Wolff,et al.  Learning Relations Using Collocations , 2001, Workshop on Ontology Learning.