Fourier Dönüşümlü Kızılötesi Spektroskopisi ile Saf ve Kızartılmış Zeytinyağının Kemometrik Öğrenme

Öz

Bu çalışmanın amacı, Fourier dönüşümlü kızılötesi spektroskopisi (FT-IR) kullanılarak karşılaştırılan saf zeytinyağı ve sekiz kez kullanılmış zeytinyağında kemometrik yöntemde kızartılmış yağların makine öğrenmesini incelemektir. Deep frying oils (DFO) 20 dakika boyunca 8 kez kullanılmıştır. Yağların kimyasal kalitesi için Cis, Trans, Ester, Metil, Karbonil, Peroksit, doymamış Peroksit ve Eter grupları incelenmiştir. Bu sonuçlar makine öğrenmesi yöntemleri kullanılarak sınıflandırma ve regresyon yapılarak değerlendirilmiştir. Bu değerlendirmeler için öncelikle bu indekslerin tüm özellikleri kullanılarak sınıflandırma ve regresyon yapılmıştır. Sınıflandırma modellerinde Destek Vektör Makineleri (SVM), K En Yakın Komşuluk (KNN), Karar Ağacı (DT) kullanılmıştır. Değerlendirme iki aşamada gerçekleştirilmiştir. İlk aşamada veri setinin yarısı eğitim, diğer yarısı da test için kullanılmıştır. İkinci aşamada, tüm veriler çapraz doğrulama (CV) yöntemi kullanılarak eğitim ve test için kullanılmıştır. Tüm veri seti kullanılarak başarı sonuçları SVM ve KNN yöntemleri ile %94 elde edilmiştir. Bu çalışmada kemometrik stratejiye göre saf zeytinyağı ile DFO arasındaki farklılıklar yüksek doğrulukla bulunmuştur. Bu olay ayrıca diğer yağ türleri ve saflık dereceleri için de mümkün olabilir. Sonuç olarak, bu yöntemin zeytinyağının bozulmasını tespit etmek için çok uygun ve doğru olduğunu göstermiştir.

Anahtar Kelimeler

• FT-IR
• KEMOMETRİK ÖĞRENME
• ZEYTİNYAĞI
• KIZARTMA YAĞI

A Novel Chemometric Learning Of Virgin And Deep Frying Olive-Oil By Fourier Transform Infrared Spectroscopy (FT-IR)

Abstract

The aim of this study is to examine the machine learning of chemometrically fried oils in virgin olive oil and eight times used olive oil compared using Fourier Transform Infrared spectroscopy (FT-IR). Deep-Frying Oils (DFO was carried out 8 times for 20 minutes. Because of chemical quality of oils, Cis, Trans, Ester, Methyl, Carbonyl, peroxide, unsatrated peroxide and ether groups were used in These results were evaluated by classification and regression using machine learning methods. For these evaluations, firstly classification and regression were made using all properties of these index. In classification models, Support Vector Machines (SVM), K Closest Neighborhood (KNN), Decision Tree (DT) were used. The evaluation was carried out in two stages. In the first stage, half of the dataset was used for training and the other half for testing. In the second stage, all data data was used for training and testing using cross validation (CV) method. The success results obtained using the all data set were 94% with Support Vector Machines and K Nearest Neighborhood methods. According to chemometric strategy, differences between virgin olive oils and DFO were found by high accurancy in this study. This phenomenon also could be possible for other oil type and degree of purity. Results illustrated that the method is very suitable and exact for detection deteriotion of olive oil.

Keywords

• FT-IR
• CHEMOMETRIC METHOD
• OLIVE OIL
• DEEP FRYING OIL

References

1. [1] V. R. Preedy and R. R. Watson, “Olives and olive oil in health and disease prevention,” Academic Press is an imprint of Elsevier, London, United Kingdom, 2021, pp. 125.
2. [2] R. Maggio, L. Cerretani, E. Chiavaro, T. Kaufman and A. Bendini, “A novel chemometric strategy for the estimation of extra virgin olive oil adulteration with edible oils,” Food Control, vol. 21, no. 6, pp. 890-895, Jun. 2010.
3. [3] D. Firestone, “Assuring the integrity of olive oil products,” Journal of AOAC International, vol. 84, no. 1, pp. 176-180, Jan. 1, 2001.
4. [4] H. T. Temiz, “Kemometrik yaklaşımlarla gıda tağşişlerinin belirlenmesinde spektroskopik yöntemlerin kullanılması,” Doktora Tezi, Fen Bilimleri Enstitüsü, Hacettepe Üniversitesi, Ankara, 2019.
5. [5] A. Çelik, “Using machine learning algorithms to detect milk quality,” Eurasian Journal of Food Science and Technology, vol. 6, no. 2, pp. 76-87, 2022.
6. [6] E. Bellou, N. Gyftokostas, D. Stefas, O. Gazeli, and S. Couris, “Laser-induced breakdown spectroscopy assisted by machine learning for olive oils classification: The effect of the experimental parameters,” Spectrochimica Acta Part B: Atomic Spectroscopy, vol. 163, no. 1, p. 105746, 2020.
7. [7] D. Stefas, N. Gyftokostas, P. Kourelias, E. Nanou, V. Kokkinos, C. Bouras, and S. Couris, “Discrimination of olive oils based on the olive cultivar origin by machine learning employing the fusion of emission and absorption spectroscopic data,” Food Control, vol. 130, no.1, p. 108318, 2021.
8. [8] N. Gyftokostas, D. Stefas, and S. Couris, “Olive oils classification via laser-induced breakdown spectroscopy,” Applied Sciences, vol. 10, no. 10, p. 3462, 2020.

9. [9] Y. Yakar, and K. Karadağ, “Identifying olive oil fraud and adulteration using machine learning algorithms,” Química Nova, vol. 45, no. 10, pp. 1245-1250, 2022.
10. [10] S. Drakopoulou, E. Orfanakis, L. Karagiannaki, F. Gaitis, S. Skoulika, A. Papaioannou, and M. Velegrakis, “Comparative evaluation of different targeted and untargeted analytical approaches to assess Greek extra virgin olive oil quality and authentication,” Molecules, vol. 27, no. 4, p. 1350, 2022.
11. [11] C. G. Viejo, and S. Fuentes, “Digital detection of olive oil rancidity levels and aroma profiles using near-infrared spectroscopy, a low-cost electronic nose and machine learning modelling,” Chemosensors, vol. 10, no. 5, p.159, 2022.
12. [12] F. Venturini, M. Sperti, U. Michelucci, I. Herzig, M. Baumgartner, J. P. Caballero, and M. A. Deriu, “Exploration of spanish olive oil quality with a miniaturized low-cost fluorescence sensor and machine learning techniques,” Foods, vol. 10, no. 5, pp.1010, 2021.
13. [13] X. Hou, G. Wang, X. Wang, X. Ge, Y. Fan, R. Jiang, and S. Nie, “Rapid screening for hazelnut oil and high‐oleic sunflower oil in extra virgin olive oil using low‐field nuclear magnetic resonance relaxometry and machine learning”, Journal of the Science of Food and Agriculture, vol. 101, no. 6, pp. 2389-2397, 2021.
14. [14] O. Gazeli, E. Bellou, D. Stefas, and S. Couris, “Laser-based classification of olive oils assisted by machine learning,” Food chemistry, vol. 302, no. 1, p. 125329, 2020.
15. [15] H. Zhao, Y. Zhan, Z. Xu, J. J. Nduwamungu, Y. Zhou, R. Powers, and C. Xu, “The application of machine-learning and Raman spectroscopy for the rapid detection of edible oils type and adulteration”, Food chemistry, vol. 373, no. 1, p. 131471, 2022.
16. [16] V. Skiada, P. Katsaris, M. E. Kambouris, V. Gkisakis, and Y. Manoussopoulos, “Classification of olive cultivars by machine learning based on olive oil chemical composition”, Food chemistry, vol. 429, no.1, p. 136793, 2023.
17. [17] B. Vega-Márquez, L. Nepomuceno-Chamorro, N. Jurado-Campos, and C. Rubio-Escudero, “Deep learning techniques to improve the performance of olive oil classification,” Frontiers in chemistry, vol. 7, 929, 2020.
18. [18] O. Gumus, E. Yasar, Z. P. Gumus, and H. Ertas, “Comparison of different classification algorithms to identify geographic origins of olive oils,” Journal of food science and technology, vol. 57, no. 1, pp. 1535-1543, 2020.
19. [19] E. Karaogul, S. Al, M. S. Karakus, A. F. Atasoy and M. H. Alma, “ Novel analyzing approaches for chemical characterization of sunflower oils during deep-frying by FT-IR,” Fresenius Environmental Bulletin and Advances in Food Sciences, vol. 29, no. 9, pp. 7847-7853, 2020.
20. [20] E. Karaogul, “Effects of asphodel tuber and dolomite on the properties of bio-hybrid films processed by a twin screw extruder,” Bioresources Journal, vol. 14, no. 2, pp. 4473-4488, April 22, 2019.
21. [21] E. Karaoğul, and M. H. Alma, “Effects of eremurus tuber and dolomite filler on several properties of poly(vinylalcohol) bio-films,” Fresenius Environmental Bulletin, vol. 28, no. 10, pp. 7108-7118, Nov. 10, 2019.
22. [22] M. Cihan, and M. Ceylan, “Comparison of Linear Discriminant Analysis, Support Vector Machines and Naive Bayes Methods in the Classification of Neonatal Hyperspectral Signatures,” Signal Processing and Communications Applications Conference (SIU), Istanbul, Turkey, 2021.
23. [23] A. Andreta, Y. Lembeye, L. L. Villa and J. C. Crébier, “Statistical modelling method for active power components based on datasheet information,” International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, Nuremberg, Germany, Jun. 1-7, 2018.
24. [24] S. A. Josephine, “Predictive Accuracy: A misleading performance measure for highly ımbalanced data classified negative,” In SAS Global Forum, pp. 942-954, 2017.
25. [25] M. Steurer, R. Hill and N. Pfeifer, “Metrics for evaluating the performance of machine learning based automated valuation models,” Journal of Property Research, vol. 38, no. 2, pp. 99-129, April 17, 2021.
26. [26] C. Ricciardi, A. S. Valente, K. Edmund, V. Cantoni, R. Green, A. Fiorillo, and M. Cesarelli, “Linear discriminant analysis and principal component analysis to predict coronary artery disease”, Health informatics journal, vol. 26, no. 3, pp. 2181-2192, 2020.
27. [27] A. Narin, Y. İşler and M. Özer, “Konjestif kalp yetmezliği teşhisinde kullanılan çapraz doğrulama yöntemlerinin sınıflandırıcı performanslarının belirlenmesine olan etkilerinin karşılaştırılması,” Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, vol. 16, no. 48, pp. 1-8, Sep.1, 2014.