Near- and Mid-Infrared Spectroscopy Combined with Machine Learning Algorithms to Determine Minerals and Antioxidant Activity in Commercial Cheese

Aug 5, 2026

Erzincan Tulum Cheese (ETC), a Protected Geographical Indication product, is evaluated for rapid quality-relevant prediction of mineral composition and antioxidant activity. Mineral content (Al, Ca, Cr, Cu, Fe, K, Mg, Mn, Na, P) is measured by ICP-MS across 70 samples, while antioxidant capacity is assessed using DPPH•+ scavenging activity. Conventional FT-NIR/FT-MIR spectra combined with PLSR show strong correlations, and information-theoretic machine learning feature selection improves results beyond PLSR. Overall, FT-NIR and FT-MIR enable non-destructive, sensitive, ~1-minute predictions for commercial cheese production.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Cheese minerals, health relevance, and labeling needs

Analytical methods and limitations

Infrared spectroscopy and machine learning approach

Turkish Journal of Agriculture-Food Science and Technology, 11(12): 2435-2445, 2023 DOI: [https://doi.org/10.24925/turjaf.v11i12.2435-2445.6526](https://doi.org/10.24925/turjaf.v11i12.2435-2445.6526) |
Turkish Journal of Agriculture-Food Science and Technology | | | --- | --- | | Available online, ISSN: 2148-127X │[www.agrifoodscience.com](www.agrifoodscience.com) │ Turkish Science and Technology Publishing (TURSTEP) | | | Near-and Mid-Infrared Spectroscopy Combined with Machine Learning
Algorithms to Determine Minerals and Antioxidant Activity in Commercial Cheese
Ahmed Menevşeoğlu1,a,*, Nurhan Güneş2,b, Huseyin Ayvaz3,c,
Sevim Beyza Özturk Sarıkaya4,d, Cuma Zehiroğlu5,e
1Department of Gastronomy and Culinary Arts, School of Tourism and Hotel Management, Agri Ibrahim Cecen University, Agri 04100, Türkiye 2Department of Electrical and Electronics Engineering, Faculty of Engineering and Natural Sciences, Sivas University of Science and
Technology, Sivas 58100, Türkiye
3Department of Food Engineering, Faculty of Engineering, Canakkale Onsekiz Mart University, Canakkale 17100, Türkiye 4Department of Food Engineering, Faculty of Engineering and Natural Sciences, Gumushane University, Gumushane 29100, Türkiye 5Scientific Research Projects Coordinatorship, Rectorate, Gumushane University, Gumushane 29100, Türkiye
*Corresponding author | | | A R T I C L E I N F O A B S T R A C T | | | Research Article | Erzincan Tulum Cheese (ETC) holds a significant place among the most popular cheeses in Türkiye. It has been awarded Protected Geographical Indication status, which restricts the allowable milk species, its production area, and specific sheep breed used in its production. Mineral content and | | Received : 04.11.2023
Accepted : 03.12.2023
Keywords:
Erzincan Tulum Cheese Minerals
FT-NIR
FT-MIR | antioxidant activity of ETC were aimed to be predicted using conventional FT-NIR and a portable FT-MIR spectrometer combined with partial least square regression (PLSR) and machine learning algorithms based on conditional entropy. Seventy ETC samples were analyzed for their mineral (Al, Ca, Cr, Cu, Fe, K, Mg, Mn, Na, and P) content using ICP-MS. The samples' antioxidant activity was measured using the DPPH•+ scavenging activity method. PLSR combined with FT-NIR spectral data correlated with antioxidant activity (r=0.89) and minerals (as low as r=0.83) except for Cr and Fe. FT-MIR data provided a good correlation for minerals (as low as r=0.82) except for Cr and Mn and a moderate correlation with antioxidant activity (r=0.64) . Information theory was | | Machine learning | applied to select wavenumbers used in machine learning algorithms, and better results were obtained compared to PLSR. Overall, FT-NIR and FT-MIR spectroscopy provided rapid (~ 1 min), non-destructive, sensitive, and reliable output for mineral and antioxidant activity predictions in | | | commercial cheese samples. | a [amenevseoglu@agri.edu.tr](amenevseoglu@agri.edu.tr) https://orcid.org/0000-0003-2454-7898 c [huseyinayvaz@comu.edu.tr](huseyinayvaz@comu.edu.tr) https://orcid.org/0000-0001-9705-6921 e [czehiroglu@gumushane.edu.tr](czehiroglu@gumushane.edu.tr) https://orcid.org/0000-0002-7185-9977 b [nurhangunes@sivas.edu.tr](nurhangunes@sivas.edu.tr) https://orcid.org/0000-0003-4163-8679 d [beyzasarikaya@gumushane.edu.tr](beyzasarikaya@gumushane.edu.tr) https://orcid.org/0000-0002-7820-4260 This work is licensed under Creative Commons Attribution 4.0 International License Introduction Cheese consumption is of great value because it consists of numerous micronutrients and trace elements. Milk and dairy products are estimated to be the primary dietary sources of calcium (Ca) and phosphorus (P), contributing to 59% and 27% of the average human daily intake, respectively. They also provide about 7% of the daily intake of Na, 9% of K, and 11% of Mg (LombardiBoccia et al., 2003) . Calcium and phosphor are essential fo

FAQs

What is the main goal of the study on Erzincan Tulum Cheese?

To predict minerals and antioxidant activity in commercial cheese using FT-NIR and FT-MIR spectroscopy combined with statistical and machine learning methods.

How were mineral contents and antioxidant activity measured for the samples?

Minerals (Al, Ca, Cr, Cu, Fe, K, Mg, Mn, Na, P) were quantified using ICP-MS, while antioxidant activity was measured with the DPPH•+ scavenging activity method.

How do the spectroscopy results compare between PLSR and machine learning?

Machine learning feature selection using information theory produced better overall results than PLSR, improving predictive performance for minerals and antioxidant activity.

Aug 5, 2026

Upload by Emma Wilson