Comparison Study of Automatic Classifiers Performance in Emotion Recognition of Arabic Social Media Users
|
Author |
Abdullah Daood, Issam Salman, Nada Ghneim |
|
Published in |
Journal of Theoretical and Applied Information Technology, Vol. 95, No. 19, 15th October 2017 |
|
Abstract |
Emotion recognition from text gained a lot of interest in the last years, but some languages such as Arabic (with its different spoken dialects) have not been given such attention. In this paper, we present our work in the Emotion detection of Arabic texts, with a focus on Levantine Twitter Messages. We have constructed a corpus of Arabic Levantine tweets, and annotated it with correspondent emotions. We implemented different methods to automatically classify text messages of individuals to infer their emotional states. We compared the results of different machine learning algorithms, and the inclusion of different features, to determine the best configuration of the emotion recognition system. Keywords: Emotional Analysis, Data Mining, Emotion Detection From Arabic Text, Twitter, Syrian Dialects. |
|
Link to read full paper |
