Öğrenci Başarısını Değerlendirme: Makine Öğrenimi ve XAI-BBO Yaklaşımının Etkisi

Abstract

Eğitimde öğrenci başarısını etkileyen faktörlerin analizi amacıyla yapılan çalışmada, veri setine çeşitli ön işleme adımları uygulanmış ve özellikle kategorik değişkenlerin etkin şekilde kullanılmasına yönelik dönüşümler uygulanmıştır. Bu dönüşümler öğrencilerin cinsiyeti, yaş aralığı, ebeveynlerin eğitim düzeyi gibi faktörleri içeriyordu. Daha sonra, en önemli 20 özelliği belirlemek için Biyocoğrafya Tabanlı Optimizasyon (BBO) algoritması kullanıldı ve bunlar daha sonra makine öğrenme modellerine dahil edildi. Değerlendirme aşamasında sonuçların elde edilmesinde Doğruluk, Hassasiyet, Geri Çağırma ve F1 puanı gibi metrikler kullanıldı. En yüksek Doğruluk değeri olan 0,7388'e Gradient Boosting algoritması ile ulaşıldı. Bu algoritmanın başarısının aydınlatılması için SHAP ve LIME yöntemleri gibi yorumlanabilir yapay zeka modelleri kullanıldı. Araştırmanın bulguları, öğrenci başarısını etkileyen faktörlerin ayrıntılı olarak incelenmesinin önemini vurgulayarak, eğitim politikalarının daha etkili bir şekilde formüle edilmesi için daha fazla araştırmaya ihtiyaç duyulduğunu vurguladı. Bu çalışmanın sonuçları eğitimde veriye dayalı karar verme süreçlerinin geliştirilmesine ve öğrenci başarısını artırmaya yönelik müdahalelerin daha etkin planlanmasına katkı sağlayabilir.

Keywords

• Açıklanabilir Yapay Zeka
• Biyocoğrafya Tabanlı Optimizasyon (BBO) algoritması
• Makine Öğrenimi modelleri
• Gradient Boosting algoritması
• SHAP yöntemi
• LIME yöntemi

Assessing Student Success: The Impact of Machine Learning and XAI-BBO Approach

Abstract

In the study conducted to analyze the factors affecting student success in education, various preprocessing steps were applied to the dataset, and transformations aimed at effectively utilizing categorical variables were particularly implemented. These transformations included factors such as students' gender, age range, and parental education level. Subsequently, the Biogeography-Based Optimization (BBO) algorithm was utilized to determine the most important 20 features, which were then incorporated into machine learning models. During the evaluation phase, metrics such as Accuracy, Precision, Recall, and F1 score were employed to obtain results. The highest Accuracy value, 0.7388, was achieved with the Gradient Boosting algorithm. To elucidate the success of this algorithm, interpretable artificial intelligence models such as SHAP and LIME methods were employed. The findings of the study underscored the importance of detailed examination of factors influencing student success, emphasizing the need for further research to formulate education policies more effectively. The results of this study may contribute to the enhancement of data-driven decision-making processes in education and the more effective planning of interventions aimed at improving student success.

Keywords

• Explalinable AI
• Biogeography-Based Optimization (BBO) algorithm
• Machine Learning models
• Gradient Boosting algorithm
• SHAP method
• LIME method

References

1. [1] Z. Akhtar, "Socio-economic status factors effecting the students achievement: a predictive study," International Journal of Social Sciences and Education, vol. 2, no. 1, pp. 281-287, 2012.
2. [2] Lakhan, G. R., Soomro, B. A., & Channa, A. (2021). INVESTIGATION OF THE SOCIO-ECONOMIC FACTORS THAT INFLUENCE YOUNG LEARNERS ACADEMIC SUCCESS: A CASE STUDY OF SECONDARY SCHOOLS OF SINDH, PAKISTAN. New Horizons (1992-4399), 15(1).
3. [3] Marks, G. N. (2016). The relative effects of socio-economic, demographic, non-cognitive and cognitive influences on student achievement in Australia. Learning and Individual Differences, 49, 1-10.
4. [4] Singh, P., & Choudhary, G. (2015). Impact of socio-economic status on academic achievement of school students: An investigation. International journal of applied research, 1(4), 266-272.
5. [5] Albashish, D., Hammouri, A. I., Braik, M., Atwan, J., & Sahran, S. (2021). Binary biogeography-based optimization based SVM-RFE for feature selection. Applied Soft Computing, 101, 107026.
6. [6] Lau, E. T., Sun, L., & Yang, Q. (2019). Modelling, prediction and classification of student academic performance using artificial neural networks. SN Applied Sciences, 1(9), 982.
7. [7] Şahin, S., & Erol, Ç. (2024). Prediction of Secondary School Students’ Academic Achievements with Machine Learning Methods and a Sample System. Cybernetics and Systems, 55(4), 940-960.
8. [8] Guleria, P., & Sood, M. (2023). Explainable AI and machine learning: performance evaluation and explainability of classifiers on educational data mining inspired career counseling. Education and Information Technologies, 28(1), 1081-1116.

9. [9] Alamri, R., & Alharbi, B. (2021). Explainable student performance prediction models: a systematic review. IEEE Access, 9, 33132-33143.
10. [10] Delen, D., Davazdahemami, B., & Rasouli Dezfouli, E. (2023). Predicting and mitigating freshmen student attrition: A local-explainable machine learning framework. Information Systems Frontiers, 1-22.
11. [11] Cruz-Jesus, F., Castelli, M., Oliveira, T., Mendes, R., Nunes, C., Sa-Velho, M., & Rosa-Louro, A. (2020). Using artificial intelligence methods to assess academic achievement in public high schools of a European Union country. Heliyon, 6(6).
12. [12] R. Hans and H. Kaur, "Hybrid Biogeography-Based Optimization and Genetic Algorithm for Feature Selection in Mammographic Breast Density Classification," International Journal of Image and Graphics, vol. 22, no. 03, p. 2140007, 2022.
13. [13] K. Bakshi and K. Bakshi, "Considerations for artificial intelligence and machine learning: Approaches and use cases," in 2018 IEEE Aerospace Conference, 2018, pp. 1-9.
14. [14] Z. Zhang, "A gentle introduction to artificial neural networks," Annals of translational medicine, vol. 4, no. 19, 2016.
15. [15] X. Y. Liew, N. Hameed, and J. Clos, "An investigation of XGBoost-based algorithm for breast cancer classification," Machine Learning with Applications, vol. 6, p. 100154, 2021.
16. [16] A. Villar and C. R. V. de Andrade, "Supervised machine learning algorithms for predicting student dropout and academic success: a comparative study," Discover Artificial Intelligence, vol. 4, no. 1, pp. 1-24, 2024.
17. [17] L. H. Alamri, R. S. Almuslim, M. S. Alotibi, D. K. Alkadi, I. Ullah Khan, and N. Aslam, "Predicting student academic performance using support vector machine and random forest," in Proceedings of the 2020 3rd International Conference on Education Technology Management, December 2020, pp. 100-107.
18. [18] H. Al-Shehri, A. Al-Qarni, L. Al-Saati, A. Batoaq, H. Badukhen, S. Alrashed, J. Alhiyafi, and S. O. Olatunji, "Student performance prediction using Support Vector Machine and K-Nearest Neighbor," in 2017 IEEE 30th Canadian Conference on Electrical and Computer Engineering (CCECE), 2017, pp. 1-4. doi: 10.1109/CCECE.2017.7946847.
19. [19] C. Bentéjac, A. Csörgő, and G. Martínez-Muñoz, "A comparative analysis of gradient boosting algorithms," Artificial Intelligence Review, vol. 54, pp. 1937-1967, 2021.
20. [20] A. Adadi and M. Berrada, "Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)," IEEE Access, vol. 6, pp. 52138-52160, 2018. doi: 10.1109/ACCESS.2018.2870052.
21. [21] I. U. Ekanayake, D. P. P. Meddage, and U. Rathnayake, "A novel approach to explain the black-box nature of machine learning in compressive strength predictions of concrete using Shapley additive explanations (SHAP)," Case Studies in Construction Materials, vol. 16, p. e01059, 2022.
22. [22] K. R. Chowdhury, A. Sil, and S. R. Shukla, "Explaining a black-box sentiment analysis model with local interpretable model diagnostics explanation (LIME)," in Advances in Computing and Data Sciences: 5th International Conference, ICACDS 2021, Nashik, India, April 23–24, 2021, Revised Selected Papers, Part I, vol. 5, pp. 90-101, Springer International Publishing, 2021.