İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ

Niyazi Kılıç

TR EN

İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ

Abstract

Bu çalışmada, obezite seviyelerinin çok sınıflı tahmininde öznitelik seçimi ve çapraz doğrulama parametrelerinin etkileri incelenmiştir. Kaggle üzerinden erişilen ve Türkiye'deki bireylerden çevrim içi anket yoluyla toplanan 1610 örnekten oluşan “Obesity” veri seti kullanılmıştır. Veri seti 14 öznitelik ve dört sınıf (zayıf, normal, fazla kilolu ve obez) içermektedir. Öznitelik seçimi amacıyla Korelasyon Tabanlı Öznitelik Seçimi yöntemi uygulanmış ve hedef değişken ile en yüksek ilişkiye sahip öznitelikler belirlenmiştir. Bu sayede öznitelik sayısı 14'ten önce 9'a, ardından 7'ye düşürülmüştür. Sınıflandırma işlemi Random Forest algoritması kullanılarak gerçekleştirilmiş ve model performansı k=2 ile k=30 arasındaki farklı k-katlı çapraz doğrulama değerleri altında sistematik olarak değerlendirilmiştir. En yüksek doğruluk değeri, 9 öznitelik kullanıldığında %93,41 olarak elde edilmiştir. Sonuçlar, öznitelik sayısının azaltılmasına rağmen sınıflandırma performansının korunduğunu ve bazı durumlarda iyileştirilebildiğini göstermektedir. Ayrıca önerilen yaklaşımın, aynı veri seti üzerinde rapor edilen çalışmalarla karşılaştırıldığında rekabetçi bir performans sergilediği görülmüştür. Bulgular, uygun öznitelik seçimi ve çapraz doğrulama stratejilerinin model karmaşıklığını azaltırken yüksek sınıflandırma başarısının elde edilmesine katkı sağlayabileceğini ortaya koymaktadır.

Keywords

MULTI-CLASS OBESITY PREDICTION USING REDUCED FEATURES AND CROSS VALIDATION STRATEGIES

Abstract

This study investigates the effects of feature selection and cross-validation parameters on multi-class obesity level prediction. The “Obesity” dataset from Kaggle, consisting of 1,610 samples collected via an online survey of individuals in Türkiye, was used. The dataset includes 14 features and four classes (underweight, normal weight, overweight, and obese). A Correlation-Based Feature Selection approach was employed to identify the features most strongly associated with the target variable. Using different correlation thresholds, two reduced feature subsets containing 9 and 7 features were generated from the original 14-feature dataset. Classification was performed using the Random Forest algorithm, and model performance was systematically evaluated under k-fold cross-validation settings ranging from k=2 to k=30. The highest classification accuracy of 93.41% was achieved using the nine-feature subset. The results demonstrate that reducing the number of features can preserve and, in some cases, improve classification performance. Furthermore, the proposed approach exhibited competitive performance compared with previously reported studies conducted on the same dataset. The findings indicate that appropriate feature selection and cross-validation strategies can reduce model complexity while maintaining high classification accuracy. These results highlight the importance of feature selection in obesity classification and suggest that simpler models with fewer features can achieve effective and reliable predictive performance.

Keywords

References

Acharjee, A., Ament, Z., West, J. A., Stanley, E., & Griffin, J. L. (2016). Integration of metabolomics, lipidomics and clinical data using a machine learning method. BMC Bioinformatics, 17(Suppl. 15), Article 440. https://doi.org/10.1186/s12859-016-1292-2
Arslan, R. U., & Yapici, I. S. (2025). A novel hybrid appRFch to enhancing obesity prediction. The European Physical Journal Special Topics, 234(15), 4843–4861. https://doi.org/10.1140/epjs/s11734-025-01620-0
Bolayır, B., Ölçer, D., Tekbıyık, B., Kara, G., & Çatuk, H. (2023). Obezite hastalığına sebep olan faktörlerin analitik hiyerarşi prosesi (AHP) yöntemiyle değerlendirilmesi. Gümüşhane Üniversitesi Sağlık Bilimleri Dergisi, 12(4), 1571–1589. https://doi.org/10.37989/gumussagbil.1366631
Börnhorst, C., et al. (2020). The role of lifestyle and non-modifiable risk factors in the development of metabolic disturbances from childhood to adolescence. International Journal of Obesity, 44, 2236–2245. https://doi.org/10.1038/s41366-020-00671-8
Canoy, D., & Buchan, I. (2007). Challenges in obesity epidemiology. Obesity Reviews, 8(Suppl. 1), 1–11. https://doi.org/10.1111/j.1467-789X.2007.00310.x
Chatterjee, A., Gerdes, M. W., & Martinez, S. G. (2020). Identification of risk factors associated with obesity and overweight—A machine learning overview. Sensors, 20, 2734. https://doi.org/10.3390/s20092734
Dugan, T. M., Mukhopadhyay, S., Carroll, A., & Downs, S. (2015). Machine learning techniques for prediction of early childhood obesity. Applied Clinical Informatics, 6, 506–520. https://doi.org/10.4338/ACI-2015-03-RA-0036
Ellis, K., Kerr, J., Godbole, S., Staudenmayer, J., & Lanckriet, G. (2016). Hip and wrist accelerometer algorithms for free-living behavior classification. Medicine and Science in Sports and Exercise, 48(5), 933–940. https://doi.org/10.1249/MSS.0000000000000840

Gutiérrez-Gallego, A., et al. (2024). Combination of machine learning techniques to predict overweight/obesity in adults. Journal of Personalized Medicine, 14, 816. https://doi.org/10.3390/jpm14080816
Karegowda, A. G., Manjunath, A. S., & Jayaram, M. A. (2010). Comparative study of attribute selection using gain ratio and correlation based feature selection. International Journal of Information Technology and Knowledge Management, 2(2), 271-277.
Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. In Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence (pp. 1137–1145).
Köklü, N., & Sulak, S. A. (2024). Using artificial intelligence techniques for the analysis of obesity status according to the individuals' social and physical activities. Sinop Üniversitesi Fen Bilimleri Dergisi, 9(1), 217–239. https://doi.org/10.33484/sinopfbd.1445215
Mert, A., Kilic, N., & Akan, A. (2011, September). Breast cancer classification by using support vector machines with reduced dimension. In Proceedings of the 53rd International Symposium ELMAR-2011 (pp. 37–40). IEEE. https://doi.org/10.1109/ELMAR.2011.6059974
Mert, A., Kılıç, N., Bilgili, E., & Akan, A. (2015). Breast cancer detection with reduced feature set. Computational and Mathematical Methods in Medicine, 2015, Article 265138. https://doi.org/10.1155/2015/265138
Pal, M. (2005). Random forest classifier for remote sensing classification. International Journal of Remote Sensing, 26(1), 217–222. https://doi.org/10.1080/01431160412331269698
Refaeilzadeh, P., Tang, L., & Liu, H. (2009). Cross-validation. In L. Liu, & M. T. Özsu (Eds.), Encyclopedia of Database Systems (pp. 532–538). Springer. https://doi.org/10.1007/978-0-387-39940-9_565
Rish, I. (2001). An empirical study of the naive Bayes classifier. In Proceedings of the IJCAI 2001 Workshop on Empirical Methods in Artificial Intelligence (pp. 41–46).
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012). An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93–104. https://doi.org/10.1016/j.isprsjprs.2011.11.002
Shi, X., et al. (2021). An ensemble-based feature selection framework to select risk factors of childhood obesity for policy decision making. BMC Medical Informatics and Decision Making, 21(1), 222. https://doi.org/10.1186/s12911-021-01580-0
Telleria-Aramburu, N., & Arroyo-Izaga, M. (2022). Risk factors of overweight/obesity-related lifestyles in university students: Results from the EHU12/24 study. British Journal of Nutrition, 127(6), 914–926. https://doi.org/10.1017/S0007114521001483
Triantafyllidis, A., et al. (2020). Computerized decision support and machine learning applications for the prevention and treatment of childhood obesity: A systematic review of the literature. Artificial Intelligence in Medicine, 104, Article 101844. https://doi.org/10.1016/j.artmed.2020.101844
Türkiye İstatistik Kurumu. (2022). Türkiye sağlık araştırması. https://data.tuik.gov.tr/ Erişim 16.02.2026.
Uğurlu, D., Yapıcı, H., Ünver, R., & Gülü, M. (2024). Comparison of obesity and physical activity levels of adult individuals by examining dietary habits with different parameters. Journal of Health Sciences and Medicine, 7(3), 301–307. https://doi.org/10.37944/jhsm.1450444
World Health Organization. (2025). Obesity and overweight. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight/ Erişim 16.02.2026.
Wu, B., Zhang, L., & Zhao, Y. (2013). Feature selection via Cramer's V-test discretization for remote-sensing image classification. IEEE Transactions on Geoscience and Remote Sensing, 52(5), 2593–2606. https://doi.org/10.1109/TGRS.2013.2263510
Yanıkkerem, E. (2017). Obezitenin Kadın Sağlığına Etkileri. Kocaeli Üniversitesi Sağlık Bilimleri Dergisi, 3(1), 37-43. https://doi.org/10.30934/kusbed.359281
Zou, J., Han, Y., & So, S. S. (2009). Overview of artificial neural networks. In D. J. Livingstone (Ed.), Artificial Neural Networks: Methods and Applications (pp. 14–22). Humana Press. https://doi.org/10.1007/978-1-60327-101-1_2
Zou, Q., Qu, K., Luo, Y., Yin, D., Ju, Y., & Tang, H. (2018). Predicting diabetes mellitus with machine learning techniques. Frontiers in Genetics, 9, Article 515. https://doi.org/10.3389/fgene.2018.00515

Details

Primary Language

Turkish

Subjects

Pattern Recognition, Machine Learning (Other), Modelling and Simulation

Journal Section

Research Article

Authors

Niyazi Kılıç ^*
0009-0007-4233-8762
Türkiye

Publication Date

September 3, 2026

Submission Date

March 24, 2026

Acceptance Date

June 29, 2026

Published in Issue

Year 2026 Volume: 29 Number: 3

IZ

https://izlik.org/JA62AR69MB

Cite

RIS / Bibtex

APA

Kılıç, N. (2026). İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 29(3), 1186-1197. https://izlik.org/JA62AR69MB

AMA

1.Kılıç N. İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ. KSU J. Eng. Sci. 2026;29(3):1186-1197. https://izlik.org/JA62AR69MB

Chicago

Kılıç, Niyazi. 2026. “İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 (3): 1186-97. https://izlik.org/JA62AR69MB.

EndNote

Kılıç N (September 1, 2026) İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 3 1186–1197.

IEEE

[1]N. Kılıç, “İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ”, KSU J. Eng. Sci., vol. 29, no. 3, pp. 1186–1197, Sept. 2026, [Online]. Available: https://izlik.org/JA62AR69MB

ISNAD

Kılıç, Niyazi. “İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29/3 (September 1, 2026): 1186-1197. https://izlik.org/JA62AR69MB.

JAMA

1.Kılıç N. İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ. KSU J. Eng. Sci. 2026;29:1186–1197.

MLA

Kılıç, Niyazi. “İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, vol. 29, no. 3, Sept. 2026, pp. 1186-97, https://izlik.org/JA62AR69MB.

Vancouver

1.Niyazi Kılıç. İNDİRGENMİŞ ÖZNİTELİKLER VE ÇAPRAZ DOĞRULAMA STRATEJİLERİ İLE ÇOK SINIFLI OBEZİTE TAHMİNİ. KSU J. Eng. Sci. [Internet]. 2026 Sep. 1;29(3):1186-97. Available from: https://izlik.org/JA62AR69MB