Machine Learning Approach to Detect Online Shopping Addiction and Study the Influencing Factors for Addiction

Abstract

Due to busy lifestyles and technological development, online shopping has grown rapidly. At the same time, the tendency to become addicted to online shopping has increased. There are significant differences between the behaviours of addicted and non-addicted people towards online shopping. The main purpose of this research is to create a machine learning model to detect this addiction and identify various e-commerce related factors that contribute to this addiction. For this research, 511 primary data were collected from online shopping users via an online survey. The questionnaire consisted of 78 questions, including their behaviour and motivation towards various features and facilities in the online shopping stores. The authors used the information gain feature ranking technique to select the most relevant features in the dataset. The models were trained using selected 11 features and 70% of data from the collected data sample. Among all the developed models' ANN showed the highest accuracy of 91%.