Sentiment analysis of shopping apps using rapidminer

An online marketplace site is a shopping place that is very popular nowadays with the community because it offers a variety of conveniences. Three of the marketplace apps are Shopee, Lazada, and Shein. Some people are satisfied with the service provided by the Shopee, Lazada, and Shein apps. However, some people complain about these applications. User-provided responses to Shopee, Lazada, and Shein apps in the comments at Google Play Store can be analyzed for negative, positive, or even neutral sentiments. This study aims to evaluate the best method to make sentiment analysis in terms of accuracy. Moreover, this study intends to identify the best shopping apps in the market based on the analysis of reviews so that customers can use the best apps easily and without any problems. As a result, Shopee, Lazada, and Shein businesses will be able to gain insight into how their customers feel about specific topics, and they will be able to identify pressing issues in real-time before they become overwhelming. The data for this study was obtained from 300 reviews we collected from the Google Play Store. After cleaning, the Naïve Bayes and SVM sentiment algorithms were used to assess the filtered summary phrases as positive or negative sentiments. According to these results, SVM is the best method because it provides significantly better accuracy than Naive Bayes for the three data pairs of Shopee, Lazada, and Shein, which are 76.00%, 78.50%, and 76.00%, respectively. It shows that SVM has outperformed Naive Bayes in terms of the accuracy of sentiment-based classification. Thus, we can conclude that Lazada is the best shopping app based on the highest accuracy achieved by SVM, which is 78.50%.