TY - GEN
T1 - Semantic sentiment analysis of Twitter
AU - Saif, Hassan
AU - He, Yulan
AU - Alani, Harith
PY - 2012
Y1 - 2012
N2 - Sentiment analysis over Twitter offer organisations a fast and effective way to monitor the publics' feelings towards their brand, business, directors, etc. A wide range of features and methods for training sentiment classifiers for Twitter datasets have been researched in recent years with varying results. In this paper, we introduce a novel approach of adding semantics as additional features into the training set for sentiment analysis. For each extracted entity (e.g. iPhone) from tweets, we add its semantic concept (e.g. Apple product) as an additional feature, and measure the correlation of the representative concept with negative/positive sentiment. We apply this approach to predict sentiment for three different Twitter datasets. Our results show an average increase of F harmonic accuracy score for identifying both negative and positive sentiment of around 6.5% and 4.8% over the baselines of unigrams and part-of-speech features respectively. We also compare against an approach based on sentiment-bearing topic analysis, and find that semantic features produce better Recall and F score when classifying negative sentiment, and better Precision with lower Recall and F score in positive sentiment classification.
AB - Sentiment analysis over Twitter offer organisations a fast and effective way to monitor the publics' feelings towards their brand, business, directors, etc. A wide range of features and methods for training sentiment classifiers for Twitter datasets have been researched in recent years with varying results. In this paper, we introduce a novel approach of adding semantics as additional features into the training set for sentiment analysis. For each extracted entity (e.g. iPhone) from tweets, we add its semantic concept (e.g. Apple product) as an additional feature, and measure the correlation of the representative concept with negative/positive sentiment. We apply this approach to predict sentiment for three different Twitter datasets. Our results show an average increase of F harmonic accuracy score for identifying both negative and positive sentiment of around 6.5% and 4.8% over the baselines of unigrams and part-of-speech features respectively. We also compare against an approach based on sentiment-bearing topic analysis, and find that semantic features produce better Recall and F score when classifying negative sentiment, and better Precision with lower Recall and F score in positive sentiment classification.
KW - sentiment analysis
KW - semantic concepts
KW - feature interpolation
UR - http://www.scopus.com/inward/record.url?scp=84868588570&partnerID=8YFLogxK
UR - http://www.springerlink.com/content/r863wj5871q17671/
U2 - 10.1007/978-3-642-35176-1-32
DO - 10.1007/978-3-642-35176-1-32
M3 - Conference publication
AN - SCOPUS:84868588570
SN - 978-3-642-35175-4
VL - 7649
T3 - Lecture notes in computer science
SP - 508
EP - 524
BT - The semantic web – ISWC 2012
A2 - Cudré-Mauroux, Philippe
A2 - Heflin, Jeff
A2 - Sirin, Evren
A2 - , et al.
PB - Springer
CY - Heildelberg (DE)
T2 - 11th international semantic web conference
Y2 - 11 November 2012 through 15 November 2012
ER -