Explainable AI-driven evaluation of plant protein rheology using tree-based and Gaussian process machine learning models
Yazarlar (7)
Mustafa Tahsin Yilmaz
Faculty Of Engineering, King Abdulaziz University, Suudi Arabistan
Faculty Of Engineering, King Abdulaziz University, Suudi Arabistan
Salman Badurayq
Industrial Area, Suudi Arabistan
Industrial Area, Suudi Arabistan
Prof. Dr. Kemal Polat Faculty Of Engineering, King Abdulaziz University, Suudi Arabistan
Ahmad H. Milyani
Center Of Excellence İn Intelligent Engineering Systems, Suudi Arabistan
Center Of Excellence İn Intelligent Engineering Systems, Suudi Arabistan
Abdulaziz S. Alkabaa
Faculty Of Engineering, King Abdulaziz University, Suudi Arabistan
Faculty Of Engineering, King Abdulaziz University, Suudi Arabistan
Prof. Dr. Osman GÜL Kastamonu Üniversitesi, Türkiye
Doç. Dr. Furkan Turker Saricaoglu Bolu Abant İzzet Baysal Üniversitesi, Türkiye
| Makale Türü |
Özgün Makale
(SSCI, AHCI, SCI, SCI-Exp dergilerinde yayınlanan tam makale)
|
||
| Dergi Adı | Ain Shams Engineering Journal (Q1) | ||
| Dergi ISSN | 2090-4479 Wos Dergi Scopus Dergi | ||
| Dergi Tarandığı Indeksler | SCI-Expanded | ||
| Makale Dili | Türkçe | Basım Tarihi | 09-2025 |
| Cilt / Sayı / Sayfa | 16 / 9 / – | DOI | 10.1016/j.asej.2025.103565 |
| Makale Linki | https://doi.org/10.1016/j.asej.2025.103565 | ||
| UAK Araştırma Alanları |
Gıda Teknolojileri
Temel İşlemler
|
||
| Özet |
| In this study, we conducted a comparative analysis of the explainability of Decision Tree Regressor (DTR) and Gaussian Process Regressor (GPR) models in predicting the shear stress and viscosity of sesame protein isolate (SPI) systems, employing explainable machine learning (EML) techniques to elucidate complex, nonlinear relationships among processing parameters. SPI samples were processed across pressure levels ranging from 0 to 100 MPa and ion concentration (IC) values from 0 to 200 mM. DTR model accurately predicted shear stress (R 2= 0.999), while a GPR model achieved high performance for viscosity prediction (R 2= 0.9925). Formally, the modeling task is framed as learning a predicting mapping function f: R p→ R, where x∈ R p denotes the vector of predictors (pressure, IC, shear rate) and y∈ R is the target variable (shear stress or viscosity), by minimizing a loss function such as mean … |
| Anahtar Kelimeler |
| Explainable artificial intelligence | Gaussian Process regressor | Sesame protein isolates | Steady shear rheology | Tree-based machine learning models |
Science Direct
Explainable AI-driven evaluation of plant protein rheology using tree-based and Gaussian process machine learning models
BM Sürdürülebilir Kalkınma Amaçları
| Atıf Sayıları | |
| Scopus | 3 |
| Google Scholar | 4 |