Profiling Of Internet Banking Users In India Using Intelligent Techniques | V. Ravi, Mahil Carr & N. Vidya Sagar

Internet banking is a new delivery channel for banks in India. The Internet banking channel is both an informative and a transactional medium. However, Internet banking has not been popularly adopted in India as expected. The objective of this paper is to find the profiles of Internet banking users as well as non-users using intelligent techniques. This study investigates and identifies potential customers based on profiles of existing users. The profiles may be used to target and attract potential customers to adopt Internet banking. Significant determining variables that influence adoption of Internet banking were identified from the literature especially from the theory of reasoned action, theory of planned behavior, technology acceptance model and diffusion of innovations theory. Likert scale responses were collected from a sample of users and non-users of Internet Banking using a questionnaire. The resultant data set was analyzed using statistical and intelligent techniques like Classification and Regression Trees (CART), Support Vector Machines (SVM), Neural Networks and Logistic Regression and classification models were built. This paper compares these four predictive models for their accuracy and usefulness. CART turns out to be a good predictive model since it also provides rules for identifying potential Internet banking users, apart from performing feature selection.