Abstract—Data conversion methods are used to transform datasets into new datasets, which may exhibit a distribution pattern different from the original dataset. Enhancing model accuracy is one of the applications of data transformation. This study compared the effectiveness of three data transformation methods: square root, logarithmic, and inverse transformation. This comparison was conducted in the context of constructing a surface roughness model for a turning process. Surface roughness plays a crucial role in determining corrosion resistance, chemical corrosion resistance, fatigue strength, and joint accuracy. These parameters significantly impact the product's operational ability and durability. An experimental turning process was performed, comprising a total of eighteen experiments designed using the Box-Behnken method. Surface roughness was selected as the response for each experiment. The three aforementioned data transformation methods were applied to the surface roughness dataset. Four surface roughness regression models were constructed, including a model without data transformation, a model with square root transformation, a model with logarithmic transformation, and a model with inverse transformation. The effectiveness of the three data transformation methods was compared using four metrics: Coefficient of Determination (R²), Adjusted Coefficient of Determination (R²(adj)), Mean Absolute Error (%MAE), and Mean Squared Error (%MSE). The study revealed that the logarithmic transformation was the most effective, followed by the square root transformation. The accuracy of the surface roughness regression model improved when utilizing these two transformations. The inverse transformation exhibited the least effectiveness among the three data transformation methods.

Keywords—square root transformation, logarithmic transformation, reciprocal transformation, surface roughness, turning

Cite: Hoang Xuan Thinh, Nguyen Truong Giang, and Vu Van Khiem, "A Comparative Analysis of Data Transformation Methods for Constructing a Surface Roughness Model in Turning Processes," International Journal of Mechanical Engineering and Robotics Research, Vol. 13, No. 4, pp. 428-434, 2024.

Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.