Title :
Emotion classification of Thai text based using term weighting and machine learning techniques
Author :
Chirawichitchai, Nivet
Author_Institution :
Fac. of Inf. Technol., Sripatum Univ., Thailand
Abstract :
In this research, I proposed Emotion Classification of Thai Text based Using Term weighting and Machine Learning Techniques focusing on the comparison of various common term weighting schemes. I found Boolean weighting with Support Vector Machine is most effective in our experiments. I also discovered that the Boolean weighting is suitable for combination with the Information gain feature selection method. The Boolean weighting with Support Vector Machine algorithm yielded the best performance with the accuracy over all algorithms. Based on our experiments, the Support Vector Machine algorithm with the Information gain feature selection yielded the best performance with the accuracy of 77.86%. Our experimental results also reveal that feature weighting methods have a positive effect on the Thai Emotion Classification Framework.
Keywords :
Boolean functions; emotion recognition; feature selection; learning (artificial intelligence); natural language processing; support vector machines; text analysis; Boolean weighting; Thai emotion classification framework; Thai text; common term weighting scheme; feature weighting method; information gain feature selection method; machine learning technique; support vector machine algorithm; term weighting technique; Emotion Classification; Feature Reduction; Machine Learning;
Conference_Titel :
Computer Science and Software Engineering (JCSSE), 2014 11th International Joint Conference on
Conference_Location :
Chon Buri
Print_ISBN :
978-1-4799-5821-4
DOI :
10.1109/JCSSE.2014.6841848