The evaluation of the hei`s entrants admission chances based on the stacking model of the support vectors machine

  1. Home
  2. Archive
  3. № 2 (63)
Author(s) Collection number Pages Download abstract Download full text
Зуб Х. В. № 2 (63) 168-176 Image Image

The choice of specialty for admission from the entrant’s point of view as one of the critical stages of the admission campaign is considered. Taking into account the large number of factors influencing the choice of entrants and the complexity of their analysis, the need to create a decision support system for entrants to higher education institutions is described. The analysis of modern domestic and foreign research concerning support of decision-making of entrants is carried out. Taking into account modern research trends, machine learning methods to predict admission to higher education institutions are chosen. To effectively solve the task of prediction, namely the high accuracy of the result, an ensemble model of machine learning methods is applied. Taking into account the results of previous authors’ research, improving the accuracy of admission to higher education institutions, chances assessment is achieved through the use of heterogeneous stacking of the Support Vectors Machine. The model is constructed based on an ensemble of four Support Vector Machine methods with different kernels: linear, polynomial, radial basis, sigmoid and logistic regression as a meta-algorithm. Classification accuracy estimates are used to analyse the results as a proportion of correctly classified examples. The obtained results indicate an increase in the accuracy of the model. The highest accuracy in all performance indicators is obtained by stacking algorithm. The obtained numerical values of the metrics are confirmed by visual analysis using ROC-curve. As a result, one can assume that the use of the chosen algorithm to solve the task of binary classification in our case has fully justified itself. The obtained numerical values of the metric are confirmed by visual analysis. A ROC curve is used to visualize the results of the study.

Keywords: admission, entrant, higher education institution, HEI, prediction, ma­chi­ne learning, ensemble learning, support vectors machine, logistic regression.

doi: 10.32403/1998-6912-2021-2-63-168-176


  • 1. Zhezhnych, P., Berezko, O., Zub, K., & Demydov, I. (2020). Analysis of Features and Abilities of Online Systems and Tools Meeting Information Needs of HEIs. Entrants, CEUR-WS.org, 2616, 76–85 (in English).
  • 2. Osadchyi, V. V., Kruhlyk, V. S., Osadcha, K. P., Serdiuk, I. M., & Bukreiev D. O. (2019). Osoblyvosti rozrobky prohramnoho zasobu dlia prohnozuvannia vstupu abituriientiv do zak­ladiv vyshchoi osvity: Vcheni zapysky Tavriiskoho Natsionalnoho universytetu imeni V. I. Ver­nadskoho. Seriia: Tekhnichni nauky, 30 (69), 1, 1 (in Ukrainian).
  • 3. Osadchyi, V. V., Kruhlyk, V. S., & Bukreiev, D. O. (2018). Rozrobka prohramnoho zasobu dlia prohnozuvannia vstupu abituriientiv do zakladiv vyshchoi osvity: Ukrainian Journal of Educational Studies and Information Technology, 6, 3. Art. no. 3 (in Ukrainian).
  • 4. Sridhar, S., Mootha, S., & Kolagati, S. (2020). A University Admission Prediction System using Stacked Ensemble Learning. 2020 Advanced Computing and Communication Tech­nologies for High Performance Applications (ACCTHPA), 162–167. doi: 10.1109/ACCTHPA49271.2020.9213205 (in English).
  • 5. Sivasangari, A., Shivani, V., Bindhu, Y., Deepa, D., & Vignesh, R. (Apr. 2021). Prediction Probability of Getting an Admission into a University using Machine Learning, in 2021 5th International Conference on Computing Methodologies and Communication (ICCMC), 1706–1709. doi: 10.1109/ICCMC51019.2021.9418279 (in English).
  • 6. AlGhamdi, A., Barsheed, A., AlMshjary, H., & AlGhamdi, H. (Mar. 2020). A Machine Lear­ning Approach for Graduate Admission Prediction, in Proceedings of the 2020 2nd International Conference on Image, Video and Signal Processing, New York, NY, USA, 155–158. doi: 10.1145/3388818.3393716 (in English).
  • 7. Sujay, S. Supervised Machine Learning Modelling & Analysis For Graduate Admission Prediction: Published in International Journal of Trend in Research and Development (IJTRD), ISSN: 2394-9333, 7 | Issue-4 , August 2020 (in English).
  • 8. Devarapalli, D. J. (2021). Classification Method to Predict Chances of Students’ Admission in a Particular College, in Proceedings of International Conference on Recent Trends in Machine Learning, IoT, Smart Cities and Applications, 1245. V. K. Gunjan and J. M. Zurada, Eds. Singapore: Springer Singapore, 225–238. doi: 10.1007/978-981-15-7234-0_19 (in English).
  • 9. Zhezhnych, P., Zub, K., Berezko, O., & Shilinh, A. (2021). A Comparison of Machine Learning Algorithms for Prediction Higher Education Institution’s Entrants Admissions, Lecture Notes on Data Engineering and Communications Technologies, 82, 171–179. doi: 10.1007/978-3-030-80475-6_17 (in English).
  • 10. Izonin, I., Trostianchyn, A., Duriagina, Z., Tkachenko, R., Tepla, T., Lotoshynska, N. (Sep. 2018). The Combined Use of the Wiener Polynomial and SVM for Material Classification Task in Medical Implants Production: International Journal of Intelligent Systems and Applications, 10, 40–47. doi: 10.5815/ijisa.2018.09.05 (in English).
  • 11. Awasthi, S. (Dec. 17, 2020). Seven Most Popular SVM Kernels. Dataaspirant. Retrieved from https://dataaspirant.com/svm-kernels/ (accessed Oct. 13, 2021) (in English).
  • 12. B. L. Ljubljana University of, Data Mining. Retrieved from https://orangedatamining.com/ (accessed Oct. 13, 2021) (in English).
  • 13. Graduate Admission 2. Retrieved from https://kaggle.com/mohansacharya/graduate-ad­mis­sions (accessed Oct. 13, 2021) (in English).