earticle

영어

This study used machine learning to derive results on how the evaluation of each function of objective consumers affects repurchase behavior. An analysis model using SVM, Rondom Forest, Decision Tree, and Neural Network, which are currently under study, was proposed, and collected reviews were divided into two categories to explore not only the leading factors of repurchase behavior but also the correlation factors. The model presented in this study confirmed that real-time collected data can be used immediately online, and that detailed positives and negatives of consumer reviews can be identified, so that user responses can be collected and utilized for decision making. In other words, it showed that practical use is possible through four algorithms: SVM, Rondom Forest, Decision Tree, and Neural Network based on machine learning. Through this, it proposed a way to use datamining- based machine learning techniques instead of collecting data through questionnaires in consumer behavior analysis.It also provided marketers with rationale and strategic implications because it could identify individual repurchase behaviors of consumers.