Filter-Based Data Partitioning for Training Multiple Classifier Systems
- 17 April 2009
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 22 (4) , 508-522
Abstract
Data partitioning methods such as bagging and boosting have been extensively used in multiple classifier systems. These methods have shown a great potential for improving classification accuracy. This study is concerned with the analysis of training data distribution and its impact on the performance of multiple classifier systems. In this study, several feature-based and class-based measures are proposed. These measures can be used to estimate statistical characteristics of the training partitions. To assess the effectiveness of different types of training partitions, we generated a large number of disjoint training partitions with distinctive distributions. Then, we empirically assessed these training partitions and their impact on the performance of the system by utilizing the proposed feature-based and class-based measures. We applied the findings of this analysis and developed a new partitioning method called "Clustering, Declustering, and Selection" (CDS). This study presents a comparative analysis of several existing data partitioning methods including our proposed CDS approach.Keywords
This publication has 29 references indexed in Scilit:
- Adaptive fusion and co-operative training for classifier ensemblesPattern Recognition, 2006
- Multistrategy ensemble learning: reducing error by combining ensemble learning techniquesIEEE Transactions on Knowledge and Data Engineering, 2004
- Input decimated ensemblesPattern Analysis and Applications, 2003
- A divide-and-conquer method for multi-net classifiersPattern Analysis and Applications, 2003
- Switching between selection and fusion in combining classifiers: an experimentIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2002
- On combining classifiersIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- Selection of relevant features and examples in machine learningArtificial Intelligence, 1997
- A Decision-Theoretic Generalization of On-Line Learning and an Application to BoostingJournal of Computer and System Sciences, 1997
- Combination of multiple classifiers using local accuracy estimatesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Diversity of Planktonic Foraminifera in Deep-Sea SedimentsScience, 1970