Analyse von Marktsegmenten mit Hilfe konnexionistischer Modelle
Abstract: Success of market segmentation depends on the use of appropriate data analysis techniques. Connectionist models (artificial neural networks) constitute an alternative to well-known methods such as linear discriminant analysis or logistic regression. So-called backpropagation networks attai...| Link(s) zu Dokument(en): | IHS Publikation |
|---|---|
| Hauptverfasser: | , |
| Format: | IHS Series NonPeerReviewed |
| Sprache: | Englisch |
| Veröffentlicht: |
Institut für Höhere Studien
1992
|
| Zusammenfassung: | Abstract: Success of market segmentation depends on the use of appropriate data analysis techniques. Connectionist models (artificial neural networks) constitute an alternative to well-known methods such as linear discriminant analysis or logistic regression. So-called backpropagation networks attain higher classification rates than logistic regression models in a pilot-study of a-priori segmentation. This may be caused by the capability of connectionist models for discovering nonlinear relationships between descriptors and segment memberships as well as interaction effects between descriptors. Special attention is drawn to specification of connectionist models and interpretation of results obtained from estimating their parameters.; |
|---|