Title :
The ROC-Boost Design Algorithm for Asymmetric Classification
Author :
Cesare, Guido ; Manduchi, Roberto
Author_Institution :
Dept. of Math., Univ. of Genova, Genova, Italy
Abstract :
In many situations (e.g., cascaded classification), it is desirable to design a classifier with precise constraints on its detection rate or on its false positive rate. We introduce ROC Boost, a modification of the Ada Boost design algorithm that produces asymmetric classifiers with guaranteed detection rate and low false positive rates. Tested in a visual text detection task, ROC-Boost was shown to perform competitively against other popular algorithms.
Keywords :
pattern classification; text detection; ROC boost; ROC-boost design algorithm; ada boost design algorithm; asymmetric classification; asymmetric classifiers; cascaded classification; false positive rates; guaranteed detection rate; visual text detection task; Algorithm design and analysis; Error analysis; Feature extraction; Histograms; Training; Upper bound; Vectors; AdaBoost; Asymmetric classifiers; ROC;
Conference_Titel :
Machine Learning and Applications and Workshops (ICMLA), 2011 10th International Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
978-1-4577-2134-2
DOI :
10.1109/ICMLA.2011.142
