Perbandingan Model Machine Learning Terbaik untuk Memprediksi Kemampuan Penghambatan Korosi oleh Senyawa Benzimidazole.

Putra Sumarjono, Cornellius Adryan; Akrom, Muhamad; Trisnapradika, Gustina Alfa

This research is an experimental study to investigate corrosion inhibitors by Benzimidazole compounds using a machine learning (ML) approach. Corrosion causes many losses arising from the loss of construction materials, work safety, and environmental pollution due to corrosion products in the form of compounds that pollute the environment. Taking an ML approach is to get the best accuracy model so that it can be used to make relevant and accurate predictions about a material. In this research, we evaluate ML algorithms with linear and non-linear methods using the k-fold cross-validation method to help measure the performance of the ML model. Referring to the coefficient of determination (R²) and root mean square error (RMSE) metrics, we conclude that the AdaBoost regressor (ADA) model is the model with the best predictive performance from the experiments we conducted from the literature for the benzimidazole compound dataset. The success of this research model offers a new perspective on the ability of ML models to predict effective corrosion inhibitors.

Techno.com, 2023, Vol 22, Issue 4, p973

1412-2693

Academic Journal

10.33633/tc.v22i4.9201