A Comparison of Online Investment Application Opinion Based on Sentiment Analysis

Abstract

As times goes by, the number of online investment applications in Indonesia continues to grow, leaving people confused about which application is best for investing. Reviews of online investment applications on the Google Play Store can provide valuable information for those looking to start investing. Analyzing application reviews requires more than just looking at star rating but also to examine the full content of review comments to understand the intent behind them. A sentiment analysis system can process reviews to extract meaningful information, including sentiment. Therefore, this study aims to compare popular online investment applications in Indonesia to determine the best one using the Naive Bayes classification. A total of 1.550 online investment app user reviews were collected to form the sample dataset. The research stages include data collection, labeling, preprocessing, classification, and evaluation. The data is divided into three categories: negative, neutral, and positive. The results show that Z application has the highest test accuracy, at 94,19%. Based on the Bag of Words analysis, the Z application is more popular among users than other applications due to its ease of use for investing.