Sentiment Analysis on User Reviews of Mutual Fund Applications

Abstract

The primary goal of this study is to compare the accuracy of the results of sentiment analysis using the Naive Bayes, Support Vector Machine (SVM), and Random Forest methods on one of the mutual fund application’s user reviews. The second goal is to identify user reviews of the mutual fund app to gain insight into the topics covered by each sentiment. The user reviews have been collected through a web scraping method on the google play store, then cleaned through several processes of data pre-processing. Feature extraction was performed using TF-IDF along with vectorization using n-grams. The model performance was measured using a confusion matrix. Using a ratio of 80:20 on training and testing data, resulting in an accuracy of 92.7, 93.7 and 94.2% for Naive Bayes, SVM, and Random Forest methods, respectively. Identify the topics covered by each sentiment in user reviews using visualizations. In the positive sentiment of users, the majority discusses the application which is easy and good, especially for novice investors. In negative sentiment, the majority discussed the slow sales process to disbursement of funds and long loading times when opening the application