Title :
Confidence based learning of a two-model committee for sequence labeling
Author :
Mancev, D. ; Todorovic, B.
Author_Institution :
Dept. of Comput. Sci., Univ. of Nis, Nis, Serbia
Abstract :
The paper presents the use of a two structural model committee, where the output of the first model together with its confidence is set as the input of the second model. The confidence for the given context of predictions in the sequence is extracted from the alternative hypotheses generated from the first model. We present experiments on the shallow parsing, comparing the performance of the proposed method to the separate models.
Keywords :
grammars; learning (artificial intelligence); confidence based learning; sequence labeling; shallow parsing; two structural model committee; Context; Hidden Markov models; Labeling; Machine learning; Predictive models; Support vector machines; Training; conditional random fields; confidence-based learning; sequence labeling; structural learning; support vector machine;
Conference_Titel :
Neural Network Applications in Electrical Engineering (NEUREL), 2012 11th Symposium on
Conference_Location :
Belgrade
Print_ISBN :
978-1-4673-1569-2
DOI :
10.1109/NEUREL.2012.6419998
