Abstract
Sentiment analysis and opinion mining in social networks present nowadays a hot topic of research. However, most of the state of the art works and researches on the automatic sentiment analysis and opinion mining of texts collected from social networks and microblogging websites are oriented towards the binary classification (i.e., classification into “positive” and “negative”) or the ternary classification (i.e., classification into “positive”, “negative” and “neutral”) of texts. In this paper, we propose a novel approach that, in addition to the aforementioned tasks of binary and ternary classification, goes deeper in the classification of texts collected from Twitter and classifies these texts into multiple sentiment classes. While in this work, we limit our scope to 7 different sentiment classes, the proposed approach is scalable and can be run to classify texts into more classes. We first introduce SENTA, our tool built to help users select out of a wide variety of features the ones that fit the most for their application, to run the classification, through an easy-to-use graphical user interface. We then use SENTA to run our own experiments of multi-class classification. Our experiments show that the proposed approach can reach up to 60.2% accuracy on the multi-class classification. Nevertheless, the approach proves to be very accurate in binary classification and ternary classification: in the former case, we reach an accuracy of 81.3% for the same dataset used after removing neutral tweets, and in the latter case, we reached an accuracy of classification of 70.1%.
| Original language | English |
|---|---|
| Journal | IEEE Access |
| DOIs | |
| Publication status | Accepted/In press - 2017 Aug 18 |
Fingerprint
Keywords
- Data mining
- Feature extraction
- Machine Learning
- Sentiment analysis
- Sentiment Analysis
- Tagging
- Tools
ASJC Scopus subject areas
- Computer Science(all)
- Materials Science(all)
- Engineering(all)
Cite this
A Pattern-Based Approach for Multi-Class Sentiment Analysis in Twitter. / Bouazizi, Mondher; Ohtsuki, Tomoaki.
In: IEEE Access, 18.08.2017.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A Pattern-Based Approach for Multi-Class Sentiment Analysis in Twitter
AU - Bouazizi, Mondher
AU - Ohtsuki, Tomoaki
PY - 2017/8/18
Y1 - 2017/8/18
N2 - Sentiment analysis and opinion mining in social networks present nowadays a hot topic of research. However, most of the state of the art works and researches on the automatic sentiment analysis and opinion mining of texts collected from social networks and microblogging websites are oriented towards the binary classification (i.e., classification into “positive” and “negative”) or the ternary classification (i.e., classification into “positive”, “negative” and “neutral”) of texts. In this paper, we propose a novel approach that, in addition to the aforementioned tasks of binary and ternary classification, goes deeper in the classification of texts collected from Twitter and classifies these texts into multiple sentiment classes. While in this work, we limit our scope to 7 different sentiment classes, the proposed approach is scalable and can be run to classify texts into more classes. We first introduce SENTA, our tool built to help users select out of a wide variety of features the ones that fit the most for their application, to run the classification, through an easy-to-use graphical user interface. We then use SENTA to run our own experiments of multi-class classification. Our experiments show that the proposed approach can reach up to 60.2% accuracy on the multi-class classification. Nevertheless, the approach proves to be very accurate in binary classification and ternary classification: in the former case, we reach an accuracy of 81.3% for the same dataset used after removing neutral tweets, and in the latter case, we reached an accuracy of classification of 70.1%.
AB - Sentiment analysis and opinion mining in social networks present nowadays a hot topic of research. However, most of the state of the art works and researches on the automatic sentiment analysis and opinion mining of texts collected from social networks and microblogging websites are oriented towards the binary classification (i.e., classification into “positive” and “negative”) or the ternary classification (i.e., classification into “positive”, “negative” and “neutral”) of texts. In this paper, we propose a novel approach that, in addition to the aforementioned tasks of binary and ternary classification, goes deeper in the classification of texts collected from Twitter and classifies these texts into multiple sentiment classes. While in this work, we limit our scope to 7 different sentiment classes, the proposed approach is scalable and can be run to classify texts into more classes. We first introduce SENTA, our tool built to help users select out of a wide variety of features the ones that fit the most for their application, to run the classification, through an easy-to-use graphical user interface. We then use SENTA to run our own experiments of multi-class classification. Our experiments show that the proposed approach can reach up to 60.2% accuracy on the multi-class classification. Nevertheless, the approach proves to be very accurate in binary classification and ternary classification: in the former case, we reach an accuracy of 81.3% for the same dataset used after removing neutral tweets, and in the latter case, we reached an accuracy of classification of 70.1%.
KW - Data mining
KW - Feature extraction
KW - Machine Learning
KW - Sentiment analysis
KW - Sentiment Analysis
KW - Tagging
KW - Tools
KW - Twitter
KW - Twitter
UR - http://www.scopus.com/inward/record.url?scp=85028513340&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85028513340&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2017.2740982
DO - 10.1109/ACCESS.2017.2740982
M3 - Article
AN - SCOPUS:85028513340
JO - IEEE Access
JF - IEEE Access
SN - 2169-3536
ER -