The Effect of Preprocessing Techniques, Applied to Numeric Features, on Classification Algorithms' Performance

It is recognized that the performance of any prediction model is a function of several factors. One of the most significant factors is the adopted preprocessing techniques. In other words, preprocessing is an essential process to generate an effective and efficient classification model. This paper investigates the impact of the most widely used preprocessing techniques, with respect to numerical features, on the performance of classification algorithms. The effect of combining various normalization techniques and handling missing values strategies is assessed on eighteen benchmark datasets using two well-known classification algorithms and adopting different performance evaluation metrics and statistical significance tests. According to the reported experimental results, the impact of the adopted preprocessing techniques varies from one classification algorithm to another. In addition, a statistically significant difference between the considered data preprocessing techniques is demonstrated.