Automatic sentiment analysis from opinion of Thais speech audio

Automatic classification of sentiment is widely used in academia and industry by several techniques. This paper aims to develop a method of sentiment analysis for Thais customers to identify the different notions into two opinions (positive or negative) to consume the products. These opinions are represented by text that is derived from the Thais speech audio content in social media especially video reviews about beauty product. Then, this work implements the model by the Naïve Bayes text classification. The results could be demonstrated that the method can provide more effectiveness and satisfactory accuracy for automatic sentiment analysis.