Sentiment Analysis on Youtube Social Media Using Decision Tree and Random Forest Algorithm: A Case Study

Abstract

People often gives opinions and thoughts on Social Media. Social media invites anyone who is interested to participate by giving feedback openly, giving comments, and sharing information in a fast and unlimited time. One of the social media that is still widely used in the community is Youtube. With so much video content, Youtube, one of which is also commonly used to promote a product. Case studies taken by researchers are public comments on Nokia’s products. Nokia was one of the best consumer stock, and was a global-scale product, but took its downfall at year 2013. By classifying positive, negative, and neutral sentiments from various opinions on Nokia’s Products, a sentiment will be conducted. From this analysis it can be determined whether the product quality is generally good or not. However, the comments are neutral labelled in dominance. The stages of sentiment analysis in this study is data preparation, data processing and evaluation. The resulting model is tested and evaluated by looking at the values of accuracy, precision, recall, and Fl-measure. The algorithm used in the conducted sentiment analysis are the Decision Tree and Random Forest Algorithm. Decision Tree algorithm has a slightly higher accuracy of 89.4% rather than Random Forest algorithm which is 88.2%.