Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/55366
Title: Mining Chinese polarity shifters
Authors: Xu, G
Huang, CR 
Issue Date: 2015
Source: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics)
Abstract: In sentiment analysis, polarity shifting means to shift the polarity of a sentiment clue. Compared with other natural language processing (NLP) tasks, to extract polarity shifters (polarity shifting patterns) in corpora is a challenging one, since the polarity shifters sometimes are very subtle, which often invalidates fully automatic approaches. In this paper, aiming to extract polarity shifters that invert or attenuate polarity, we use a semi-automatic approach based on pattern mining. The approach can greatly reduce the human annotating cost and cover as many polarity shifters as possible. We tested this approach on domain corpora, and encouraging experimental results are reported.
Keywords: Polarity shifting
Prior polarity
Sentiment analysis
Sequence mining
Publisher: Springer
Journal: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics) 
ISBN: 9783319271934
ISSN: 0302-9743
EISSN: 1611-3349
DOI: 10.1007/978-3-319-27194-1_25
Description: 16th Workshop, CLSW 2015, Beijing, China, May 9-11, 2015
Appears in Collections:Conference Paper

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

115
Last Week
6
Last month
Citations as of Sep 28, 2020

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.