Sentiment Monitoring of Google Maps Reviews Using Naive Bayes and Support Vector Machine at Aveta Hotel Malioboro

Abstrak

Manual monitoring of hotel customer reviews on Google Maps is increasingly inefficient due to growing review volumes, particularly in detecting negative sentiment. This study proposes a web-based sentiment monitoring system integrated with Naive Bayes and Support Vector Machine (SVM) for automated sentiment classification of Google Maps reviews at Aveta Hotel Malioboro. A dataset of 2,449 historical reviews was collected via DOM scraping and processed through text preprocessing and TF-IDF feature extraction. SVM with minority-class oversampling achieved the highest performance: accuracy 0.9540, precision 0.9514, recall 0.9540, and F1 Score 0.9522. The implemented system retrieves reviews periodically via SerpAPI, classifies sentiment automatically, and delivers real-time Telegram notifications, enabling structured and data-driven hotel review monitoring

Identifier

• OpenAlex: https://openalex.org/W7167247479
• DOI: https://doi.org/10.55927/fjst.v5i6.95