Sign in Subscribe
Algorithms

Comparative Performance of Sentiment Classification Algorithms on SIGNAL Reviews

Strategic Insights

— 1 min read

Sriwijaya University researchers have investigated the comparative performance of four algorithms in analyzing the sentiment of user feedback obtained from the Google Play Store. The study focused on user reviews of the public service app SIGNAL, which provides public access to information and vehicle tax payments. The research team evaluated the performance of Naive Bayes, Random Forest, Decision Tree, and SVM algorithms in sentiment analysis, using a dataset of 36,000 user feedback. The findings indicate that Random Forest performed the best overall performance, followed by Naive Bayes, SVM, and Decision Tree.

Key Takeaways:

  • The study examined the comparative performance of four algorithms (Naive Bayes, Random Forest, Decision Tree, and SVM) in analyzing the sentiment of user feedback obtained from the Google Play Store.
  • The dataset consisted of 36,000 user reviews from the Google Play Store, which underwent preprocessing, feature extraction using TF-IDF, and class balancing using SMOTE.
  • The model evaluation was conducted using accuracy, precision, recall, and F1-score metrics.
  • Random Forest performed the best overall performance (accuracy 91.04%, F1-score 94.80%), followed by Naive Bayes (accuracy 89.89%, F1-score 93.38%), SVM (accuracy 89.22%, F1-score 93.02%), and Decision Tree (accuracy 88.40%, F1-score 92.31%).
  • The findings indicate that Random Forest is highly effective for balanced datasets, while SVM and Naive Bayes offer competitive precision for applications prioritizing accuracy in positive class detection.
  • The output of this study can be applied practically by developers and related institutions in optimizing public service applications.

Statistics:

  • 36,000 user feedback obtained from the Google Play Store were analyzed.
  • The dataset underwent preprocessing, feature extraction using TF-IDF.
  • Class balancing was performed using SMOTE.
  • Model evaluation was conducted using accuracy (91.04% for Random Forest), precision, recall, and F1-score (94.80% for Random Forest).

Sources:

  • "Performance Comparison of Sentiment Classification Algorithms on SIGNAL Reviews Using SMOTE." Journal of Information Systems and Informatics, 2025,7(3):2407-2423.
  • Journal of Information Systems and Informatics, https://doi-org.sdpl.idm.oclc.org/10.51519/journalisi.v7i3.1196

Read more