1993 | OriginalPaper | Buchkapitel
Application of Neural Networks to Gradient Search Techniques in Cluster Analysis
verfasst von : Stephane Delsert, Denis Hamad, Mohamed Daoudi, Jack-Gérard Postaire
Erschienen in: Artificial Neural Nets and Genetic Algorithms
Verlag: Springer Vienna
Enthalten in: Professional Book Archive
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this paper, we propose an approach to cluster analysis based on two levels: a neural network and an heuristic clustering algorithm. The neural network provides data compression such as the probability density function (p.d.f.) of the weights must be as near as to the p.d.f. of the data. During the clustering step which is based on the estimation of the gradient of the p.d.f., each available vector weight, or prototype, is moved in the direction of the p.d.f. gradient approximation. The process is iterated until the normalised local gradient is equal to zero, which result near the modes of the p.d.f.