Diagnosis with Topological Maps

  • This technical report is a compilation of several papers on the task of solving diagnostic problems with the help of topology preserving maps. It first reviews the application of Kohonen's Self- Organizing Feature Map (SOFM) for a technical diagnosis task, namely the fault detection in CNC-Machines with the KoDiag system [RW93], [RW94]. For emergent problems with coding attribute values, we then introduce fuzzy coding, similarity assignment and weight updating schemes for three crucial data types (continuous values, ordered and unordered symbols). These techniques result in a SOFM type network based on user defined local similarities, thus being able to incorporate a priori knowledge about the domain [Rah95].

Export metadata

  • Export Bibtex
  • Export RIS

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Jürgen Rahmel
URN (permanent link):urn:nbn:de:hbz:386-kluedo-599
Serie (Series number):LSA Report (95,9E)
Document Type:Preprint
Language of publication:English
Year of Completion:1999
Year of Publication:1999
Publishing Institute:Technische Universität Kaiserslautern
Tag:diagnostic problems ; topology preserving maps
Faculties / Organisational entities:Fachbereich Informatik
DDC-Cassification:004 Datenverarbeitung; Informatik

$Rev: 12793 $