Analysing user trust in electronic banking using data mining methods

The potential fraud problems, international economic crisis and the crisis of trust in markets have affected financial institutions, which have tried to maintain customer trust in many different ways. To maintain these levels of trust they have been forced to make significant adjustments to economic structures, in efforts to recoup their investments and maintain the loyalty of their customers. To achieve these objectives, the implementation of electronic banking for customers has been considered a successful strategy. The use of electronic banking in Spain in the last decade has been fostered due to its many advantages, giving rise to real integration of channels in financial institutions. This paper reviews different methods and techniques to determine which variables could be the most important to financial institutions in order to predict the likely levels of trust among electronic banking users including socio-demographic, economic, financial and behavioural strategic variables that entities have in their databases. To do so, the most recent advances in machine learning and soft-computing have been used, including a new selection operator for multiobjective genetic algorithms. The results obtained by the algorithms were validated by an expert committee, ranking the quality of them. The new methodology proposed, obtained the best results in terms of optimisation as well as the highest punctuation given by the experts.