Analisis Sentimen Tingkat Kepuasan Aplikasi WordPress Menggunakan Metode K-Nearest Neighbor dan Naive Bayes

Moch Siddiq Hamid (Universitas Muhammadiyah Sidoarjo, Indonesia)
Ade Eviyanti (Universitas Muhammadiyah Sidoarjo, Indonesia)
Hindarto Hindarto (Universitas Muhammadiyah Sidoarjo, Indonesia)
Novia Ariyanti (Universitas Muhammadiyah Sidoarjo, Indonesia)

Smatika Jurnal : STIKI Informatika Jurnal
Vol 15 No 01 (2025): SMATIKA Jurnal : STIKI Informatika Jurnal

User satisfaction reflects emotions when comparing services received with expectations, so understanding user satisfaction is important for app development. This research aims to evaluate user satisfaction with WordPress apps on the Google Play Store and identify areas for improvement. Sentiment analysis with KNN and Naïve bayes algorithms as the method used to extract information from 5,000 user reviews downloaded from Google Play Store,. The results showed the majority of reviews had positive sentiments, with Naïve Bayes providing better results than KNN, achieving 88% accuracy, 89.45% precision, 88% recall, and 83% F1-Score on a 90:10 data split. The word cloud of positive reviews featured words such as “great”, “good”, “helpful”, “app”, and “good”, reflecting user satisfaction with the ease and benefits of the app, while negative reviews featured words such as “difficult”, “try”, and “fail” indicating technical difficulties and user dissatisfaction. This study concludes that WordPress apps have provided a satisfactory experience for most users, but some technical areas need improvement. The results of this study will provide valuable information for app developers in efforts to improve service quality and the app's reputation