Dipl.-Ing. Sylvia Melzer

- externe Doktorandin -

 

Institut für Informationssysteme
Universität zu Lübeck
Ratzeburger Allee 160 ( Gebäude 64 - 2.OG )
D-23562 Lübeck


Öffnet ein Fenster zum Versenden einer E-Mail E-Mail an Sylvia Melzer
 
 

Semantic Assets: Latent Structures for Knowledge Management

In standard information retrieval systems, queries can be specified with differ- ent languages (string patterns, logical formulas, and so on). It is well known that it is hard to simultaneously maximize quality measures for query answers, such as, e.g., precision and recall.

Retrieval of documents with high recall, while maintaining at least a decent precision level, is a frequent problem in knowledge management (KM) contexts based on information retrieval (IR) processes, e.g., for information association in business contexts. Similar to making implicit knowledge explicit, deriving explicit symbolic content descriptions is important in KM tasks.

In this thesis we show how explicit symbolic descriptions can be combined with implicit holistic content representations known from information retrieval in order to support knowledge management processes in general, and informa- tion association based on IR query answering in particular. The methodology exhibited in this thesis is verified using representative examples, and validated with feasibility and effectiveness studies.

Publications

  • S. Espinosa, A. Kaya, S. Melzer, R. Möller, and M. Wessel. Towards a Media Interpretation Framework for the Semantic Web. The 2007 IEEE/ WIC/ ACM International Conference on Web Intelligence (WI‘07), IEEE Computer Society, Washington, DC, USA, pages 374-380, 2007 (pdf)
  • S. Espinosa, A. Kaya, S. Melzer, and R. Möller. On Ontology Based Abduction for Text Interpretation. Proceedings of 9th International Conference on Intelligent Text Processing and Computational Linguistics (Ed.: A. Gelbukh), number 4919 in LNCS, Springer, pages 194-205, 2008 (pdf)
  • S. Melzer. On the Relationship between Ontology-based and Holistic Representations in a Knowledge Management System. Ontology-based Applications for Enterprise Systems & Knowledge Management (Eds.: M. Nazir Ahmad, R. Colomb, and M. Abdullah). IGI Global, pages 292-323, 2013 (pdf)
  • S. Melzer, R. God, T. Kiehl, R. Möller, and M. Wessel. Identifikation von Varianten durch Berechnung der semantischen Differenz von Modellen. Tagungsband zum Tag des Systems Engineering (Eds.: M. Maurer, S.O. Schulze), Carl Hanser Verlag, pages 279-288, 2014 (slidespdf)
  • T. Bahns, S. Melzer, R. God, and D. Krause. Ein modellbasiertes Vorgehen zur variantengerechten Entwicklung modularer Produktfamilien, Tagungsband zum Tag des Systems Engineering (Hrsg.: Chr. Muggeo, S.O. Schulze), Carl Hanser Verlag, pages 141-150, 2015 (slidespdf)
  • S. Melzer, U. Wittke, H. Hintze, and R. God.  Physische Architekturen variantengerecht aus Funktionalen Architekturen für Systeme (FAS) spezifizieren, Tagungsband zum Tag des Systems Engineering (Eds.: S.O. Schulze, C. Tschirner, R. Kaffenberger, S. Ackva), Carl Hanser Verlag, pages 429-438, 2016 (slidespdf)
  • D. Arndt, S. Melzer, R. God, and M. Sieber. Konzept zur Verhaltensmodellierung mit der Systems Modeling Language (SysML) zur Simulation varianten Systemverhaltens. Tagungsband zum Tag des Systems Engineering (Eds.: S.O. Schulze, C. Tschirner, R. Kaffenberger, S. Ackva), Carl Hanser Verlag, pages 115-124, 2017 (slidespdf)