Title :
Research on Dynamic Integration Algorithm Based on Local Learning Strategy
Author :
Ying Junping ; Wan Yuan
Author_Institution :
Art & Media Coll., Ningbo Dahongying Univ., Ningbo, China
Abstract :
Taking local learning strategy as the local member classifier generating means, local performance estimation of classifier as in-depth analysis method of classifier, this paper, systematically investigates dynamic integration method based on local learning strategy and local performance estimation, on the basis of improving the integrated classifier accuracy with efficiency consideration in order to obtain superior performance of the dynamic integration classification algorithm.
Keywords :
learning (artificial intelligence); pattern classification; dynamic integration classification algorithm; in-depth analysis method; integrated classifier accuracy; local learning strategy; local member classifier; local performance estimation; Accuracy; Classification algorithms; Estimation; Heuristic algorithms; Learning systems; Machine learning; Training; dynamic integration algorithm; local learning strategy;
Conference_Titel :
Intelligent Systems (GCIS), 2012 Third Global Congress on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4673-3072-5
DOI :
10.1109/GCIS.2012.78