期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：1
页码：292
DOI：10.15680/IJIRCCE.2017.0501043
出版社：S&S Publications
摘要：In this paper, introduce a joint segmentation as well as classification framework for audio sentimentanalysis. Past sentiment classification algorithms divides a sentence as a word sequence, which does not effectivelyhandle the inconsistent sentiment polarity between a phrase and the words which is obtained by audio file. It may faceproblem in case if it contains words such as not bad and a great deal of. We address this problem by developing a jointsegmentation as well as classification framework (JSC). These frameworks simultaneously conduct sentencesegmentation as well as sentence-level sentiment classification which are present in audio file. The joint model istrained only based on the annotated sentiment polarity of sentences present in audio, without any segmentationannotations.
关键词：Artificial intelligence; joint segmentation and classification; natural language processing; sentiment;analysis; sentiment classification.Audio data segmentation and classification.
