DM: Questions on neural nets for data mining

To what extent have artificial neural networks been used successfully for
data mining? The weights of a trained multilayer perceptron are difficult
to interpret, and one may have to resort to some type of rule extraction
procedure in order to obtain a summary of the input-output mapping provided
by the network. Hybrid neuro-fuzzy systems are intrinsically easier in this
respect.

Visualization is an aspect of data mining, and neural nets have been used
successfully for this purpose, but are they any better than other
visualization techniques, for example, independent components analysis?

Richard


-------------------------------
Richard Dybowski PhD
Research Fellow (Knowledge & Data Engineering)
King's College London

Tel (office): (0)20 7928 9292 extension 6429
Tel (mobile): 079 76 25 00 92
Fax: +44 (0)20 7928 4458
E-mail: richard.dybowski@kcl.ac.uk
Web site: http://www.umds.ac.uk/microbio/richard/richard.htm

• Prev by Date: DM: Exceptions Identification
• Next by Date: DM: CFP: Learning Statistical Models from Relational Data, AAAI 2000
• Prev by thread: DM: CFP: Learning Statistical Models from Relational Data, AAAI 2000
• Next by thread: DM: Exceptions Identification
• Index(es):
  • Date
  • Thread