In this study, a clustering-based sales forecasting scheme by combining K-means algorithm with SVR and three
similarity measurements is proposed. Although there are many different clustering algorithms have been proposed,
but K-means algorithm is still one of the most effective clustering algorithms [11]. This study considers the SVR as
the predictor due to its great potential and superior performance in practical applications [13]. SVR based on
statistical learning theory is an effective neural network algorithm for solving nonlinear regression estimation
problems and has been successfully used in sales forecasting[6-8].