APPLICATION OF THE RANDOM FOREST ALGORITHM IN PREDICTING IMPULSE BUYING BY CONSUMERS ON INDONESIAN MARKETPLACE APPLICATIONS

Ni Nyoman Cory Villdina (Universitas Bina Sarana Informatika)
Monica Julia Efata (Universitas Bina Sarana Informatika)
Etania Frederika (Universitas Bina Sarana Informatika)
Muhammad Sony Maulana (Universitas Bina Sarana Informatika)

Jurnal Pilar Nusa Mandiri
Vol. 22 No. 1 (2026): Pilar Nusa Mandiri : Journal of Computing and Information System Publishing Pe

The development of e-commerce in Indonesia has driven an increase in impulsive buying behavior influenced by emotional and situational factors. This study aims to determine the important factors that influence impulsive buying by applying the Random Forest machine learning algorithm. Data were obtained from 628 online shopping application users aged 18-40 years in Indonesia. The analysis results indicated that the majority of respondents fall into the medium impulsive buying category (67.9%), with a model accuracy of  86.5%. The Hedonic Browsing (HB) variable had the greatest influence on impulsive buying behavior, followed by Utilitarian Browsing (UB) and Time Perspective (TP). These findings confirm that emotional aspects and situational conditions play a more dominant role than rational considerations in driving spontaneous shopping decisions.