Klasifikasi Ulasan Pengguna Tiket Pesawat Online dengan Penanganan Ketidakseimbangan Data Menggunakan SMOTE dengan Machine Learning

Husaini, Rahayun
Amrullah Husaini, Rahayun
Pratama, Gede Yogi
Satrani, Azral

Jurnal Tata Kelola dan Kerangka Kerja Teknologi Informasi
Vol. 11 No. 3 (2025): Desember 2025

The COVID-19 pandemic affected public habits in air travel and increased the use of online ticket booking platforms. This study aimed to analyze sentiment in online flight ticket purchase reviews using the Support Vector Machine and K-Nearest Neighbor methods. The research was conducted by collecting user review data from the Tiket.com website, followed by preprocessing, term weighting using TF-IDF, and classification using both methods. The results show that the Support Vector Machine method achieves an accuracy of 51 percent, while the K-Nearest Neighbor method reaches 55 percent after applying data balancing techniques. This study concludes that both methods are effective in classifying user sentiment and can assist service providers in improving service quality and understanding customer needs