Performance Comparison of New Adjusted Min-Max with Decimal Scaling and Statistical Column Normalization Methods for Artificial Neural Network Classification

In this research, the normalization performance of proposed adjusted min-max methods was compared to statistical column, decimal scaling, and methods, in terms accuracy mean square error final classification outcomes. The evaluation process employed an artificial neural network on a large variety widely used datasets. best method normalization, providing 84.0187% average ranking 0.1097 across all six However, adjusted-2 achieved higher lower than each following datasets: white wine quality, Pima Indians diabetes, vertical Indian liver disease For example, quality dataset 100% 0.00000282 error. To conclude, for some applications one these specific datasets, should be over other tested because it performed better.