<![CDATA[Online job search applications are proliferating and are crucial for job seekers in Indonesia. As seen in Google Play Store reviews, KitaLulus, a leading platform, faces technical issues, unresponsive services, and limited job postings. This study analyzes user sentiment using the Naive Bayes algorithm. Data was collected from 1,000 reviews through web scraping between September and November 2024. The pre-processing steps included text cleaning, tokenization, stopword removal, and stemming. It classified reviews into positive, neutral, and negative sentiments. A confusion matrix evaluated the model using accuracy, precision, recall, and F1-score. Results showed positive reviews, but some users reported performance issues and limited features. The Naive Bayes model achieved 88% accuracy, 87% precision, 88% recall, and an 85% F1 score. This method efficiently processes extensive text data with lower computational costs than KNN and SVM. This research helps improve application development, enhance service quality, and expand sentiment analysis studies in IT. The findings will guide the creation of innovative strategies to benefit the community.]]>
Copyrights © 2025
