Social media is a medium where people can express their opinion on something. Opinion mining or sentiment analysis, which is studying people's sentiments towards certain entities. This can be used by companies to find out people's responses to a sales product. Sentiment analysis has received a lot of attention in recent years. Sentiment analysis is one of the main tasks of NLP (Natural Language Processing). In this paper, sentiment polarity categorization becomes the basis for sentiment analysis problems in product reviews. A general process for sentiment polarity categorization is proposed with a detailed description of the process. The data used in this study is an online product review collected from the Tokopedia application. Classification is carried out on sentence level categorization and star rating level categorization. There are three models used to compare the classification process, namely SVM, Random Forest, and Naïve Bayes models. The results of this research paper are in the form of a comparison of the performance of the three models against the polarity categorization of product review sentiment on Tokopedia
Copyrights © 2022