Description: RSE (Regularized Selective Ensemble) is a selective ensemble learning algorithm for binay classification, which constructs ensemble under the regularization framework. In current version, the graph Laplacian serves as the regularizer, and unlabeled data can also be exploited to improve the performance.
References: N. Li and Z.-H. Zhou. Selective ensemble under regularization framework. In: Proceedings of the 8th International Workshop on Multiple Classifier Systems (MCS'09), Reykjavik, Iceland, LNCS 5519, 2009, pp.293-303.
ATTN: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact Prof. Zhi-Hua Zhou (zhouzh@nju.edu.cn).
Requirement:
- To use this package, the hole WEKA environment must be available. This package is developed with WEKA 3.5 and MATLAB 2008a. For more information about WEKA, Please refer: I.H. Witten and E. Frank. Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations. Morgan Kaufmann, San Francisco, CA, 2000.
- Please make sure that WEKA is available in MATLAB. To do this, you should add weka.jar to the classpath of MATLAB.
- MOSEK is recommend to the QP problem in RSE, since it is much faster than the default QP solver in Matlab. You can just add the MOSEK optimization toolbox into the path of MATLAB. For more information, please refer to MOSEK Optimization toolbox for MATLAB.
ATTN2: This package was developed by Mr. Nan Li. For any problem concerning the code, please feel free to contact Mr. Nan Li (lin@lamda.nju.edu.cn).
Download:
code (23Kb)