TY - GEN
T1 - Reliability assessment of ensemble classifiers
T2 - 9th International Workshop on Digital Mammography, IWDM 2008
AU - Mazurowski, MacIej A.
AU - Zurada, Jacek M.
AU - Tourassi, Georgia D.
PY - 2008
Y1 - 2008
N2 - In classifier ensembles predictions of different classifiers regarding a query are combined into one final decision. It was previously shown that using ensemble techniques can significantly improve classification performance. In this study we build upon this result and propose to use variability in the predictions of classifiers contributing to the final decision as an indicator of its reliability. The study hypothesis is tested with respect to previously proposed information-theoretic computer-aided decision (IT-CAD) system for detection of masses in mammograms. A database of 1820 regions of interest (ROIs) extracted from digital database of screening mammography (DDSM) is used. Experimental results show that the proposed reliability assessment successfully identifies decisions that can not be trusted. Further, a low correlation between reliability and the classifier output is noted. This opens a possibility of combining reliability and ensemble output into one improved decision.
AB - In classifier ensembles predictions of different classifiers regarding a query are combined into one final decision. It was previously shown that using ensemble techniques can significantly improve classification performance. In this study we build upon this result and propose to use variability in the predictions of classifiers contributing to the final decision as an indicator of its reliability. The study hypothesis is tested with respect to previously proposed information-theoretic computer-aided decision (IT-CAD) system for detection of masses in mammograms. A database of 1820 regions of interest (ROIs) extracted from digital database of screening mammography (DDSM) is used. Experimental results show that the proposed reliability assessment successfully identifies decisions that can not be trusted. Further, a low correlation between reliability and the classifier output is noted. This opens a possibility of combining reliability and ensemble output into one improved decision.
UR - https://www.scopus.com/pages/publications/50949099927
U2 - 10.1007/978-3-540-70538-3_51
DO - 10.1007/978-3-540-70538-3_51
M3 - Conference contribution
AN - SCOPUS:50949099927
SN - 3540705376
SN - 9783540705376
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 366
EP - 370
BT - Digital Mammography - 9th International Workshop, IWDM 2008, Proceedings
Y2 - 20 July 2008 through 23 July 2008
ER -