Artificial Neural Network Approach for Modelling Overall Customer Satisfaction

M.H. Askariazad and M. Wanous (UK)

Keywords

Prediction with Expert Advice, Artificial Neural Network, Modelling, Customer Satisfaction, and Online questionnaire.

Abstract

A three layer Artificial Neural Network (ANN) model was developed to predict the overall customer satisfaction considering 10 different satisfaction dimensions (criteria) based on the knowledge of Iranian marketing experts. An on-line questionnaire was conducted to collect the Marketing and Sales Managers’ expectation of their customer satisfaction. After backpropagation (BP) training the initial model was able to predict the overall satisfaction with ten neurons in input layer, five neurons in hidden layer and one neuron in output layer. Levenberg-Marquardt (LM) algorithm was found as the best of 11 BP algorithms with the minimum Mean Square Error (MSE) in both training and testing data sets. The optimal ANN structure was investigated by increasing the number of neurons and hidden layers. Based on the defined Performance Index (PI) the optimal model was selected. The linear regression between the selected network outputs and the corresponding targets were proven to be satisfactory with a correlation coefficient of 0.826. The study is particularly useful for marketers in evaluating the overall customer satisfaction using different satisfaction dimensions. The results may act as a general view of experts towards customer satisfaction in the Iranian context and can be used as a guideline for the companies implementing customer satisfaction practices.

Important Links:



Go Back