Türk Lise Öğrencilerinin Mesleki Eğilimlerini Tahmin Etme: Farklı Makine Öğrenmesi Yöntemlerinin Yeni Bir Eğitim Verisi Üzerinde Karşılaştırılması

Abstract

21. yüzyılda makine öğrenmesi (MÖ) uygulamalarının kullanıldığı alanların sayısı giderek artmaktadır. Eğitim, MÖ kullanımının son zamanlarda yaygınlaştığı alanlardan biridir. Günümüzde eğitim sistemindeki veriler devasa boyutlara ulaşmıştır. Bu verilerden anlamlı veri setleri oluşturulabilir ve bu veri setleri MÖ yöntemleri yardımıyla işlenebilir. Bununla birlikte, birçok çalışmada kullanılan veri setleri, öğrenciler veya öğretmenlerle yapılan ve tartışmalı/öznel veriler içeren anketlere dayanmaktadır. Bu çalışmada ise, 2014-2020 yılları arasında mezun olan öğrencilerin ders notları, mezuniyet yılları ve cinsiyet bilgilerini içeren 4 yıllık bir lisenin gerçek verilerinden yeni bir veri seti oluşturulmaktadır. Veri setindeki özellikler ile bu öğrencilerin kazandıkları üniversite bölümleri arasındaki ilişkiler dikkate alınmakta ve veriler üzerinde MÖ modelleri eğitilerek öğrencilerin mesleki eğilimlerinin tahmin edilmesi amaçlanmaktadır. Öğrencilerin kazandığı bölümlerin yer aldığı veri setinde sınıflandırma için çok sayıda farklı bölüm olduğundan, ilgili bölümleri ve meslekleri aynı sınıflandırma etiketi altında gruplandıran bir etiket gruplama yöntemiyle 3 ek veri seti daha oluşturulmaktadır. Sonrasında bu veri setleri için farklı MÖ modelleri eğitilerek analiz edilmektedir. Ayrıca, bu modellerden başarılı sonuç üreten 6 tanesi ile bir çoğunluk oylama yöntemi önerilmektedir. Sonuç olarak performans değerlendirmesi için; eğitim doğruluğu, Cohen puanı, F1 puanı ve test doğruluğu olmak üzere 4 farklı değerlendirme kriteri kullanılmaktadır. Bulgular, oluşturulan yeni eğitim veri seti için farklı MÖ yöntemleri ele alınarak sunulmakta ve tartışılmaktadır. Elde edilen bulgulara göre, klasik makine öğrenmesi yöntemleri kullanıldığında %75’e kadar tahmin doğruluğu elde edilirken, bu makine öğrenmesi yöntemleri kullanılarak geliştirilen çoğunluk oylama yöntemiyle başarım %80’e yaklaşmaktadır.

Keywords

• Veri seti oluşturma
• eğitim
• makine öğrenmesi
• mesleki eğilim tahmini

Vocational Tendency Prediction of Turkish High School Students: Comparison of Different Machine Learning Methods on a Novel Educational Data

Abstract

In the 21st century, the number of areas where machine learning (ML) applications have emerged is gradually increasing. Education is one of the areas where the use of ML has spread, recently. The data in the education system has reached gigantic sizes. Hence, meaningful datasets can be created and processed with the help of ML methods. However, many studies still consult for questionnaires made on students or teachers that produce controversial and subjective data. In this study, a novel dataset is created from real-world data of a 4-year high school, including course grades, graduation years and gender information of the students graduated between 2014-2020. It is aimed to predict vocational tendency of students by designing ML models on past data considering the relationships between the attributes in the dataset and the university departments enrolled. Moreover, since there is a huge number of different departments for classification, 3 additional datasets are also constructed by a label grouping method that groups related departments and professions into the same classification label. Then, different ML models are trained on these datasets and analyzed. Additionally, a majority voting method is proposed by using 6 of these models producing successful results. As a result, the performance of these methods is evaluated according to 4 different evaluation criteria, including training accuracy, Cohen score, F1 score and test accuracy. The findings are presented and discussed by addressing the availability of different ML methods for such a novel educational dataset constructed. According to the findings, a prediction accuracy is achieved up to 75% when classical machine learning methods are applied, and the prediction approaches 80% with the majority voting method constructed by using these machine learning methods.

Keywords

• Dataset construction
• education
• machine learning
• vocational tendency prediction

References

1. Delors, J. (1996b). Learning : the treasure within : report to UNESCO of the International Commission on Education for the Twenty-first Century. http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=007405168&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA
2. Kucak, D., Juricic, V., & Dambic, G. (2018). Machine Learning in Education - A survey of current research Trends. In Annals of DAAAM for & proceedings of the . International DAAAM Symposium (pp. 0406–0410). https://doi.org/10.2507/29th.daaam.proceedings.059
3. Ocaña-Fernández, Y., Valenzuela-Fernández, L. A., & Garro-Aburto, L. L. (2019). Artificial Intelligence and Its Implications in Higher Education. Propositos Y Representaciones, 7(2), 553–568. https://doi.org/10.20511/pyr2019.v7n2.274
4. Raja, R., & Nagasubramani, P. C. (2018). Impact of modern technology in education. Journal of Applied and Advanced Research, 3(1),33–35. https://doi.org/10.21839/jaar.2018.v3is1.165
5. Alpaydin, E. (2016b). Machine Learning : The new AI. https://bvbr.bib-bvb.de:443/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=029233302&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA
6. Pedro, F., Subosa, M., Rivas, A., & Valverde, P. (2019). Artificial intelligence in education: Challenges and opportunities for sustainable development. (UNESCO Publication No ED-2019/WS/8). https://unesdoc.unesco.org/ark:/48223/pf0000366994
7. Roll, I., & Wylie, R. (2016). Evolution and revolution in artificial intelligence in education. International Journal of Artificial Intelligence in Education, 26(2), 582–599. https://doi.org/10.1007/s40593-016-0110-3
8. Timms, M. J. (2016). Letting artificial intelligence in education out of the box: educational cobots and smart classrooms. International Journal of Artificial Intelligence in Education, 26(2), 701–712. https://doi.org/10.1007/s40593-016-0095-y

9. Holmes, W., Bialik, M., & Fadel, C. (2019b). Artificial intelligence in Education: Promises and implications for teaching and learning. In Open Research Online (The Open University).
10. Luckin, R. (2018). Machine Learning and Human Intelligence: The Future of Education for the 21st century. In CERN Document Server (European Organization for Nuclear Research). http://cds.cern.ch/record/2698126
11. Savaş, S. (2021). Artificial intelligence and innovative applications in education: the case of Turkey. Journal of Information Systems and Management Research, 3(1), 14–26. https://dergipark.org.tr/tr/pub/jismar/issue/63377/852043
12. Nafea, I. T. (2018). Machine learning in educational technology. In InTech eBooks. https://doi.org/10.5772/intechopen.72906
13. Anand, V. K., Rahiman, S. K. A., George, E. B., & Huda, A. S. (2018). Recursive clustering technique for students’ performance evaluation in programming courses. 2018 Majan International Conference (MIC), 1–5. https://doi.org/10.1109/mintc.2018.8363153
14. Salloum, S. A., Salloum, A., & Alfaisal, R. (2024). Objectives and obstacles of artificial intelligence in education. In Studies in big data (pp. 605–614). https://doi.org/10.1007/978-3-031-52280-2_38
15. Darji, J., & Tulsidas Nakrani. (2021). Machine Learning Based Prediction Technique for Student’s Performance. 7(6), 8. https://www.researchgate.net/publication/354330470_Machine_Learning_Based_Prediction_Technique_for_Student
16. Chen, L., Chen, P., & Lin, Z. (2020). Artificial Intelligence in Education: a Review. IEEE Access, 8(8), 75264–75278. https://doi.org/10.1109/ACCESS.2020.2988510
17. Alam, M. M., Mohiuddin, K., Das, A. K., Islam, M. K., Kaonain, M. S., & Ali, M. H. (2018). A Reduced feature based neural network approach to classify the category of students. Proceedings of the 2nd International Conference on Innovation in Artificial Intelligence, 28–32. https://doi.org/10.1145/3194206.3194218
18. Hasan, R., Ovy, M. K. A., Nishi, I. Z., Hakim, M. A., & Hafiz, R. (2020). A Decision Support System of Selecting Groups (Science/ Business Studies/ Humanities) for Secondary School Students in Bangladesh. 2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 1–6. https://doi.org/10.1109/icccnt49239.2020.9225411
19. Demir, M. K. (2016). Problems encountered by school principals: Unchanging facts of changing Turkey. The Anthropologist, 23(3), 629–640. https://doi.org/10.1080/09720073.2014.11891983
20. Dayioğlu, M., & Türüt-Aşik, S. (2007). Gender differences in academic performance in a large public university in Turkey. Higher Education, 53(2), 255–277. https://doi.org/10.1007/s10734-005-2464-6
21. Swanzen, R. (2018). Facing the generation chasm: The parenting and teaching of generations Y and Z. International Journal of Child, Youth and Family Studies, 9(2), 125–150.
22. Dolot, A. (2018). The characteristics of Generation Z. E-mentor, 74, 44–50. https://doi.org/10.15219/em74.1351
23. Pokhrel, S., & Chhetri, R. (2021). A Literature Review on Impact of COVID-19 Pandemic on Teaching and Learning. Higher Education for the Future, 8(1), 133–141. https://doi.org/10.1177/2347631120983481
24. Çelik, E. F., & Özçevik, Y. (2022). Feasibility of machine learning methods for digital transformation on high school education. 6(7), 7. https://www.researchgate.net/publication/363813826_FEASIBILITY_OF_MACHINE_LEARNING_METHODS_FOR_DIGITAL_TRANSFORMATION_ON_HIGH_SCHOOL_EDUCATION
25. Probst, P., Wright, M. N., Boulesteix, A., Probst, P., Wright, M. N., & Boulesteix, A. (2019). Hyperparameters and tuning strategies for random forest. Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery, 9(3). https://doi.org/10.1002/widm.1301
26. Wazirali, R. (2020). An improved intrusion detection system based on knn hyperparameter tuning and cross-validation. Arabian Journal for Science and Engineering, 45(12), 10859–10873. https://doi.org/10.1007/s13369-020-04907-7
27. Rojas-Dominguez, A., Padierna, L. C., Valadez, J. M. C., Puga-Soberanes, H. J., & Fraire, H. J. (2017). Optimal Hyper-Parameter tuning of SVM classifiers with application to medical diagnosis. IEEE Access, 6, 7164–7176. https://doi.org/10.1109/access.2017.2779794
28. Feng, Y., Wang, D., Yin, Y., Li, Z., & Hu, Z. (2020). An XGBoost-based casualty prediction method for terrorist attacks. Complex & Intelligent Systems, 6(3), 721–740. https://doi.org/10.1007/s40747-020-00173-0
29. Dev, V. A., & Eden, M. R. (2019). Formation lithology classification using scalable gradient boosted decision trees. Computers & Chemical Engineering, 128, 392–404. https://doi.org/10.1016/j.compchemeng.2019.06.001
30. Shankar, K., Zhang, Y., Liu, Y., Wu, L., & Chen, C. (2020). Hyperparameter tuning deep learning for diabetic retinopathy fundus image classification. IEEE Access, 8, 118164–118173. https://doi.org/10.1109/access.2020.3005152
31. Çılgın, C., Gökçen, H., & Gökşen, Y. (2022). Sentiment analysis of public sensitivity to COVID-19 vaccines on twitter by majority voting classifier-based machine learning. Journal of the Faculty of Engineering and Architecture of Gazi University, 38(2), 1093–1104. https://doi.org/10.17341/gazimmfd.1030198
32. Halabaku, E., & Bytyçi, E. (2024). Overfitting in machine learning: A comparative analysis of decision trees and random forests. Intelligent Automation & Soft Computing, 39(6), 987–1006. https://doi.org/10.32604/iasc.2024.059429