TY - GEN
T1 - Corpus construction on polarity shifting in sentiment analysis
AU - Li, Shoushan
AU - Lee, Yat Mei
AU - Huang, Chu-ren
PY - 2013/12/1
Y1 - 2013/12/1
N2 - Polarity shifting has been a challenge to automatic sentiment classification. In this paper, we create a corpus which consists of polarity-shifted sentences in the product reviews, where both the sentimental words and shifting trigger words are annotated. In particular, we group the polarity-shifted sentence structures into five main categories, i.e., negation, contrastive transition, modality, implication, and irrelevance Evaluation shows the statistics on the agreement of the annotation and the distribution of the five categories of polarity shifting is given.
AB - Polarity shifting has been a challenge to automatic sentiment classification. In this paper, we create a corpus which consists of polarity-shifted sentences in the product reviews, where both the sentimental words and shifting trigger words are annotated. In particular, we group the polarity-shifted sentence structures into five main categories, i.e., negation, contrastive transition, modality, implication, and irrelevance Evaluation shows the statistics on the agreement of the annotation and the distribution of the five categories of polarity shifting is given.
UR - http://www.scopus.com/inward/record.url?scp=84893346018&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-45185-0_65
DO - 10.1007/978-3-642-45185-0_65
M3 - Conference article published in proceeding or book
SN - 9783642451843
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 625
EP - 634
BT - Chinese Lexical Semantics - 14th Workshop, CLSW 2013, Revised Selected Papers
T2 - 14th Workshop on Chinese Lexical Semantics, CLSW 2013
Y2 - 10 May 2013 through 12 May 2013
ER -