Investigation of Corrosion Inhibition Efficiency of Pyridine-Quinoline Compounds through Machine Learning

Journal of Multiscale Materials Informatics, Journal Year: 2024, Volume and Issue: 1(1), P. 1 - 9

Published: April 29, 2024

Corrosion in materials is a significant concern for the industrial and academic fields because corrosion causes enormous losses various such as economy, environment, society, industry, security, safety, others. Currently, material damage control using organic compounds has become popular field of study. Pyridine quinoline stand out inhibitors among myriad they are non-toxic, inexpensive, effective variety corrosive environments. Experimental investigations developing candidate potential inhibitor time resource-intensive. In this work, we use quantitative structure-property relationship (QSPR)-based machine learning (ML) approach to investigate support vector (SVR), random forest (RF), k-nearest neighbors (KNN) algorithms predictive models inhibition performance. (Inhibition efficiency) pyridine-quinoline derivative on iron. We found that RF model showed best ability based coefficient determination (R2) root mean squared error (RMSE) metrics. Overall, our study provides new insights regarding ML predicting iron surfaces.

Language: Английский

Investigasi Model Machine Learning Regresi Pada Senyawa Obat Sebagai Inhibitor Korosi

Jurnal Algoritma, Journal Year: 2024, Volume and Issue: 21(1), P. 332 - 342

Published: July 29, 2024

Korosi merupakan tantangan signifikan bagi daya tahan material, yang seringkali menyebabkan kerugian ekonomi besar. Penelitian ini memanfaatkan teknik Machine Learning (ML) untuk memprediksi efektivitas senyawa obat sebagai inhibitor korosi. Kami menggunakan lima algoritma ML menonjol: Regresi Linear, Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Random Forest, dan XGBoost. Model-model dilatih dievaluasi dataset terdiri dari 14 fitur molekuler dengan efisiensi inhibisi korosi (IE%) variabel target. Hasil pelatihan model awal mengidentifikasi Forest XGBoost berkinerja terbaik berdasarkan metrik seperti Mean Squared Error (MSE), Root (RMSE), Absolute (MAE), R-squared (R²). Penyetelan hiperparameter lebih lanjut GridSearchCV menunjukkan bahwa XGBoost, setelah penyetelan, secara mengungguli lainnya, mencapai kesalahan terendah nilai R² tertinggi, akurasi prediktif superior aplikasi ini. Temuan menegaskan potensi ML, khususnya dalam meningkatkan pemodelan korosi, sehingga memberikan wawasan berharga bidang ilmu