EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Comparative Study of Machine Learning Algorithms for Sentiment Classification on Genshin Impact User Reviews

Farida Arindra, Agung Febrian

JASMINE: Journal of Intelligent Systems and Machine Learning · 28.09.2026 · 1 (2) · 142-151

DOI: 10.25124/jasmine.11238 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JASMINE: Journal of Intelligent Systems and Machine Learning, online erschienen am 28.09.2026. Verfasst von Farida Arindra und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mobile game user reviews contain valuable insights regarding gameplay experience, technical performance, and overall player satisfaction, yet manually analyzing large volumes of textual feedback is inefficient. This study presents a comparative sentiment classification analysis of Genshin Impact user reviews collected from the Google Play Store to identify the most effective machine learning model for sentiment analysis in the mobile gaming domain. A total of 22,255 review records were obtained from a publicly available Kaggle dataset, followed by data cleaning, binary sentiment labeling based on rating scores, and text preprocessing including normalization, stopword removal, and lemmatization. Textual features were extracted using Term Frequency-Inverse Document Frequency (TF-IDF), and four machine learning algorithms, namely Multinomial Naive Bayes, Logistic Regression, Linear Support Vector Machine, and XGBoost, were evaluated under identical experimental conditions. Performance was assessed using accuracy, precision, recall, F1-score, Macro F1-score, and ROC-AUC metrics. The results showed that Logistic Regression achieved the best overall performance, with 88.10% accuracy, 91.19% F1-score, and 94.13% ROC-AUC. The findings demonstrate that classical machine learning approaches remain highly effective for sentiment classification in mobile gaming review environments and provide practical insights for developers to analyze player feedback and improve user experience.

Zitieren

APA

Arindra, F., & Febrian, A. (2026). A Comparative Study of Machine Learning Algorithms for Sentiment Classification on Genshin Impact User Reviews. JASMINE: Journal of Intelligent Systems and Machine Learning, 1(2), 142-151. https://doi.org/10.25124/jasmine.11238

Vancouver

Arindra F, Febrian A. A Comparative Study of Machine Learning Algorithms for Sentiment Classification on Genshin Impact User Reviews. JASMINE: Journal of Intelligent Systems and Machine Learning. 2026;1(2):142-151. doi:10.25124/jasmine.11238

BibTeX
@article{Arindra2026,
  author = {Arindra, Farida and Febrian, Agung},
  title = {A Comparative Study of Machine Learning Algorithms for Sentiment Classification on Genshin Impact User Reviews},
  journal = {JASMINE: Journal of Intelligent Systems and Machine Learning},
  year = {2026},
  volume = {1},
  number = {2},
  pages = {142--151},
  doi = {10.25124/jasmine.11238}
}
Kostenlos starten →