Analisis Sentimen Masyarakat terhadap Keamanan Penggunaan E-Commerce B2C Menggunakan Pendekatan Naïve Bayes Berbasis Text Mining untuk Mencegah Penipuan

Marcelena Vicky Galena (Universitas Airlangga, Indonesia)
Adnan Syawal Adilaha Sadikin (Universitas Airlangga, Indonesia)
Aprilia Prastyaningrum (Universitas Airlangga, Indonesia)
Reza Febrian Nugroho (Universitas Airlangga, Indonesia)
M. Fariz Fadillah Mardianto (Universitas Airlangga, Indonesia)

G-Tech : Jurnal Teknologi Terapan
Vol 8 No 3 (2024): G-Tech, Vol. 8 No. 3 Juli 2024

The Society 5.0 Era's technology development has shifted marketing communication from face-to-face to screen interaction, like online shopping via e-commerce. According to Statista Market Insight data, e-commerce users in Indonesia reached 178.94 million in 2022, with transactions totaling Rp476.3 trillion. Despite its growth, e-commerce is prone to cybercrime, with 16,845 reports to the National Police's Ditipideksus from 2017 to 2020. This research analyzes public sentiment on e-commerce security through Play Store and App Store comments. The Naïve Bayes model shows an accuracy of 80% on the Play Store and 87% on the App Store, with AUCs of 0.864 and 0.942, respectively, indicating excellent sentiment classification performance. The findings aim to help B2C e-commerce providers enhance security through advanced technologies, user education, fraud detection systems, and improved transparency and response to security incidents, thereby increasing user trust.