Title :
Convex Hull Ensemble Machine
Author_Institution :
Dept. of Stat., Ewha Woman´´s Univ., Seoul, South Korea
Abstract :
We propose a new ensemble algorithm called Convex Hull Ensemble Machine (CHEM). CHEM in Hilbert space is developed first and it is modified to regression and classification problems. Empirical studies show that in classification problems CHEM has similar prediction accuracy as AdaBoost, but CHEM is much more robust to output noise. In regression problems, CHEM works competitively with other ensemble methods such as Gradient Boost and Bagging.
Keywords :
Hilbert spaces; data mining; decision trees; learning (artificial intelligence); pattern classification; statistical analysis; AdaBoost; Bagging; CHEM; Convex Hull Ensemble Machine; Gradient Boost; Hilbert space; classification; data mining; decision trees; ensemble algorithm; machine learning; output noise; regression; Accuracy; Bagging; Decision trees; Geometry; Hilbert space; Machine learning; Machine learning algorithms; Noise robustness; Solid modeling; Statistics;
Conference_Titel :
Data Mining, 2002. ICDM 2003. Proceedings. 2002 IEEE International Conference on
Print_ISBN :
0-7695-1754-4
DOI :
10.1109/ICDM.2002.1183909
