Classifying emotion in Thai youtube comments

Author

Sarakit, Phakhawat ; Theeramunkong, Thanaruk ; Haruechaiyasak, Choochart ; Okumura, Manabu

Author_Institution

Sch. of ICT, Thammasat Univ., Pathumthani, Thailand

fYear

2015

fDate

22-24 March 2015

Firstpage

1

Lastpage

5

Abstract

To add more value on YouTube, a popular portal of social media clips, it is worth recognizing automatically the mood of a media clip using the comments given to such clip. This paper presents a method to classify emotion of a Thai media clip on YouTube using the comments given to the clip. Six basic emotions considered are Anger, Disgust, Fear, Happiness, Sadness and Surprise. Performances using three alternative machine learning algorithms, namely multinomial naïve Bayes (MNB), decision tree (DT) and support vector machine (SVM) are compared. As the result, SVM achieves the highest accuracy in the commercial advertisement (AD) genre with an accuracy of 76.14% while MNB with yields the best result in the music video (MV) genre with an accuracy of 84.48%.

Keywords

Bayes methods; decision trees; emotion recognition; human computer interaction; learning (artificial intelligence); pattern classification; portals; social networking (online); support vector machines; AD genre; DT; MNB; MV genre; SVM; Thai YouTube comments; Thai media clip; anger; commercial advertisement genre; decision tree; disgust; emotion classification; fear; happiness; machine learning algorithms; media clip mood; multinomial naïve Bayes; music video genre; portal; sadness; social media clips; support vector machine; surprise; Accuracy; Conferences; Media; Sentiment analysis; Support vector machines; Twitter; YouTube; Emotion Classification; Emotion Detection; Machine Learning; Text Analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Information and Communication Technology for Embedded Systems (IC-ICTES), 2015 6th International Conference of

Conference_Location

Hua-Hin

Type

conf

DOI

10.1109/ICTEmSys.2015.7110808

Filename

7110808