An indexing approach for representing multimedia objects in high-dimensional spaces based on expectation maximization algorithm
Contributo in Atti di convegno
Data di Pubblicazione:
2005
Citazione:
An indexing approach for representing multimedia objects in high-dimensional spaces based on expectation maximization algorithm / G. Boccignone, V. Caggiano, C. Cesarano, V. Moscato, L. Sansone - In: Advances in Multimedia Information Systems / [a cura di] K. Selcuk, A. Celentano. - [s.l] : Springer, 2005. - ISBN 3540287922. - pp. 63-77 (( Intervento presentato al 11. convegno International Workshop on Advances in Multimedia Information Systems tenutosi a Sorrento nel 2005.
Abstract:
In this paper we introduce a new indexing approach to representing multimedia object classes generated by the Expectation Maximization clustering algorithm in a balanced and dynamic tree structure. To this aim the EM algorithm has been modified in order to obtain at each step of its recursive application balanced clusters. In this manner our tree provides a simple and practical solution to index clustered data and support efficient retrieval of the nearest; neighbors in high dimensional object spaces.
Tipologia IRIS:
03 - Contributo in volume
Keywords:
EM algorithm
Elenco autori:
G. Boccignone, V. Caggiano, C. Cesarano, V. Moscato, L. Sansone
Link alla scheda completa:
Titolo del libro:
Advances in Multimedia Information Systems