| Titre : |
Le Clustering des données : une nouvelle approche évolutionnaire quantique |
| Type de document : |
texte imprimé |
| Auteurs : |
Chafika Ramdane ; Univ. de Constantine, Éditeur scientifique ; S. Chikhi, Directeur de thèse |
| Année de publication : |
2006 |
| Importance : |
126 f. |
| Note générale : |
01 Disponible à la salle de recherche 02 Disponibles au magazin de la B.U.C. 01 CD |
| Catégories : |
Français - Anglais
Informatique
|
| Tags : |
Clustering Données Approche évolutionnaire quantique |
| Index. décimale : |
004 Traitement de données. Informatique |
| Résumé : |
In this work, we deal with the problem of data clustering. The clustering represents a fu
ndamental task for a great number of different fields. It is a very current method which allows a better understanding of the analyzed data set. This problem can be modelled as a problem of optimization. For its resolution, we have developed a new quantum evolutionary approach based on a quantum representation to code the space of search and quantum evolutionary strategy of search to optimize a measure of quality of cluster in order to find a good partitioning of data set. The particularity of our approach is the reduced size of the population and the reasonable number of iterations to find the best partitioning of data due to quantum operations involved in the search, which are based on principles of quantum computing as the superposition of the states, the interference and others. Another particularity lies in the initialization step where new important function is used. The other characteristic is the capacity of this approach to being extended in a simple way. Results on synthetic and real data sets are very promising and show the capacity and the efficiency of the approach to be identified valid clusters of various densities, various sizes and various forms as well as the independence between the final partit ion found and the initial partition. |
| Diplôme : |
Magistère |
| En ligne : |
../theses/informatique/RAM4510.pdf |
| Permalink : |
https://bu.umc.edu.dz/md/index.php?lvl=notice_display&id=227 |
Le Clustering des données : une nouvelle approche évolutionnaire quantique [texte imprimé] / Chafika Ramdane ; Univ. de Constantine, Éditeur scientifique ; S. Chikhi, Directeur de thèse . - 2006 . - 126 f. 01 Disponible à la salle de recherche 02 Disponibles au magazin de la B.U.C. 01 CD
| Catégories : |
Français - Anglais
Informatique
|
| Tags : |
Clustering Données Approche évolutionnaire quantique |
| Index. décimale : |
004 Traitement de données. Informatique |
| Résumé : |
In this work, we deal with the problem of data clustering. The clustering represents a fu
ndamental task for a great number of different fields. It is a very current method which allows a better understanding of the analyzed data set. This problem can be modelled as a problem of optimization. For its resolution, we have developed a new quantum evolutionary approach based on a quantum representation to code the space of search and quantum evolutionary strategy of search to optimize a measure of quality of cluster in order to find a good partitioning of data set. The particularity of our approach is the reduced size of the population and the reasonable number of iterations to find the best partitioning of data due to quantum operations involved in the search, which are based on principles of quantum computing as the superposition of the states, the interference and others. Another particularity lies in the initialization step where new important function is used. The other characteristic is the capacity of this approach to being extended in a simple way. Results on synthetic and real data sets are very promising and show the capacity and the efficiency of the approach to be identified valid clusters of various densities, various sizes and various forms as well as the independence between the final partit ion found and the initial partition. |
| Diplôme : |
Magistère |
| En ligne : |
../theses/informatique/RAM4510.pdf |
| Permalink : |
https://bu.umc.edu.dz/md/index.php?lvl=notice_display&id=227 |
|
Affiner la recherche Générer le flux rss de la recherche
Partager le résultat de cette recherche Interroger des sources externesClassification automatique des données avec erreurs de mesure d'application à un jeu de données médicales / Amar Rebbouh
