Ce livre examine la nature des ensembles de données déséquilibrés et examine deux méthodes externes, qui peuvent augmenter les performances d'un apprenant sur des classes sous-représentées. Les deux techniques équilibrent artificiellement les données d'entraînement ; l'une en rééchantillonnant aléatoirement des exemples de la classe sous-représentée et en les ajoutant à l'ensemble d'entraînement, l'autre en supprimant au hasard des exemples de la classe surreprésentée de l'ensemble d'entraînement. Un schéma de combinaison est ensuite présenté. L'approche est celle dans laquelle plusieurs classificateurs sont disposés dans une structure hiérarchique en fonction de leurs techniques d'échantillonnage. L'architecture se compose de deux experts, l'un qui augmente les performances en combinant des classificateurs qui rééchantillonnent les données d'entraînement à des rythmes différents, l'autre en combinant des classificateurs qui suppriment les données des données d'entraînement à des vitesses différentes. En utilisant la mesure F, qui combine précision et rappel en tant que statistique de performance, le schéma de combinaison est efficace pour apprendre des ensembles de données gravement déséquilibrés. En fait, par rapport à une technique de combinaison de pointe, Adaptive-Boosting, le système proposé s'avère supérieur pour l'apprentissage sur des ensembles de données déséquilibrés.
Les informations fournies dans la section « Synopsis » peuvent faire référence à une autre édition de ce titre.
Vendeur : BuchWeltWeit Ludwig Meier e.K., Bergisch Gladbach, Allemagne
Taschenbuch. Etat : Neu. This item is printed on demand - it takes 3-4 days longer - Neuware -This book investigates the nature of imbalanced data sets and looks at two external methods, which can increase a learner s performance on under represented classes. Both techniques artificially balance the training data; one by randomly re-sampling examples of the under represented class and adding them to the training set, the other by randomly removing examples of the over represented class from the training set. A combination scheme is then presented. The approach is one in which multiple classifiers are arranged in a hierarchical structure according to their sampling techniques. The architecture consists of two experts, one that boosts performance by combining classifiers that re-sample training data at different rates, the other by combining classifiers that remove data from the training data at different rates. Using the F-measure, which combines precision and recall as a performance statistic, the combination scheme is shown to be effective at learning from severely imbalanced data sets. In fact, when compared to a state of the art combination technique, Adaptive-Boosting, the proposed system is shown to be superior for learning on imbalanced data sets. 220 pp. Englisch. N° de réf. du vendeur 9783639762211
Quantité disponible : 2 disponible(s)
Vendeur : moluna, Greven, Allemagne
Etat : New. Dieser Artikel ist ein Print on Demand Artikel und wird nach Ihrer Bestellung fuer Sie gedruckt. Autor/Autorin: Rashid Syed ZahidurThe author s research interests are in the areas of machine learning, data mining, information acquisition, and decision theory. Specifically, in active learning, active inference, interactive machine learning, sta. N° de réf. du vendeur 151400880
Quantité disponible : Plus de 20 disponibles
Vendeur : buchversandmimpf2000, Emtmannsberg, BAYE, Allemagne
Taschenbuch. Etat : Neu. This item is printed on demand - Print on Demand Titel. Neuware -This book investigates the nature of imbalanced data sets and looks at two external methods, which can increase a learner¿s performance on under represented classes. Both techniques artificially balance the training data; one by randomly re-sampling examples of the under represented class and adding them to the training set, the other by randomly removing examples of the over represented class from the training set. A combination scheme is then presented. The approach is one in which multiple classifiers are arranged in a hierarchical structure according to their sampling techniques. The architecture consists of two experts, one that boosts performance by combining classifiers that re-sample training data at different rates, the other by combining classifiers that remove data from the training data at different rates. Using the F-measure, which combines precision and recall as a performance statistic, the combination scheme is shown to be effective at learning from severely imbalanced data sets. In fact, when compared to a state of the art combination technique, Adaptive-Boosting, the proposed system is shown to be superior for learning on imbalanced data sets.VDM Verlag, Dudweiler Landstraße 99, 66123 Saarbrücken 220 pp. Englisch. N° de réf. du vendeur 9783639762211
Quantité disponible : 1 disponible(s)
Vendeur : AHA-BUCH GmbH, Einbeck, Allemagne
Taschenbuch. Etat : Neu. nach der Bestellung gedruckt Neuware - Printed after ordering - This book investigates the nature of imbalanced data sets and looks at two external methods, which can increase a learner s performance on under represented classes. Both techniques artificially balance the training data; one by randomly re-sampling examples of the under represented class and adding them to the training set, the other by randomly removing examples of the over represented class from the training set. A combination scheme is then presented. The approach is one in which multiple classifiers are arranged in a hierarchical structure according to their sampling techniques. The architecture consists of two experts, one that boosts performance by combining classifiers that re-sample training data at different rates, the other by combining classifiers that remove data from the training data at different rates. Using the F-measure, which combines precision and recall as a performance statistic, the combination scheme is shown to be effective at learning from severely imbalanced data sets. In fact, when compared to a state of the art combination technique, Adaptive-Boosting, the proposed system is shown to be superior for learning on imbalanced data sets. N° de réf. du vendeur 9783639762211
Quantité disponible : 1 disponible(s)