METHOD OF THE ANALYSIS OF FOREIGN LANGUAGE KNOWLEDGE LEVEL OF THE STUDENTS OF HIGHER EDUCATION ESTABLISHMENT ON THE BASE OF MACHINE LEARNING

Authors

  • Kozachko Oleksii Vinnytsia National Technical University
  • Zhukov Serghiy Vinnytsia National Technical University
  • Vuzh Tetiana Vinnytsia M. I. Pirogov National Medical University
  • Lototskyi Andriy Vinnytsia National Technical University

DOI:

https://doi.org/10.31649/2307-5392-2021-4-13-20

Keywords:

decision tree, intelligence analysis, models construction, revealing of regularities, impact factor, Python

Abstract

Paper is devoted to the development of the method of the analysis of the level of foreign language knowledge of the students of higher education establishment on the base of machine learning methods. By means of machine learning methods the regularities and trends may be determined which enable to improve the level of knowledge of the students of higher education establishments. The tasks of the paper, according to aim, put forward is versatile analysis of the data: analysis of the subject area, intelligence analysis, model construction and determination of the subjects, influencing the level of foreign language study. For the search of the type of theclassifier mathematical model the decision trees are used in the given research.

For the development of the method of the analysis of thelevel of  foreign language knowledge  of the  students the technologies of machine learning are used, by means of these technologies various models of decision trees are developed with further selection of the best one. Regularities determination is carried out as a result of decision trees construction on the samples of grades, obtained by the students of М. І. PirogovVinnytsiaNational Medical University in the 2nd, 4th and 6th semesters. Within the frame of this study the search of the classifier type was carried out on the base of gradient boosting and logistic regression. The experiments, carried out, showed that the rules, obtained by means of regression model more accurately forecast the level of foreign language knowledge. On the base of these studies adequate and rather accurate conclusions regarding the revealed regularities are mode.

The suggested method enables to reveal the regularities of determination of the level of foreign language knowledge of the students of higher education establishments, using the methods of machine learning and determine subjects, having greatest impact of the level of foreign language knowledge. For the determination of the level of foreign language knowledge of  the students of higher education establishments the program module in the form of the web-system, using the basic web-technologies was developed, the module allows to solve the task, put forward, applying automatic facilities and provide recommendations, regarding the improvement  of foreign language knowledge. Program module comprises the web-site with the connected data base. The results of the research can be efficiently used for the improvement of modern education process.

Author Biographies

Kozachko Oleksii , Vinnytsia National Technical University

Cand. Sc. (Eng.), Assistant Professor with the Department of System Analysis and Information Technologies

Zhukov Serghiy, Vinnytsia National Technical University

Cand. Sc. (Eng.), Assistant Professor with the Department of System Analysis and Information Technologies

Vuzh Tetiana, Vinnytsia M. I. Pirogov National Medical University

Cand. Sc. (Eng.), Assistant Professor with the Department of Biological Physics, Medical Equipment and Informatics

Lototskyi Andriy, Vinnytsia National Technical University

Master Student with the Department of System Analysis and Information Technologies

Downloads

Abstract views: 101

Published

2022-10-12

How to Cite

[1]
O. . Kozachko, S. Zhukov, T. Vuzh, and A. Lototskyi, “METHOD OF THE ANALYSIS OF FOREIGN LANGUAGE KNOWLEDGE LEVEL OF THE STUDENTS OF HIGHER EDUCATION ESTABLISHMENT ON THE BASE OF MACHINE LEARNING”, Works of VNTU, no. 4, Oct. 2022.

Issue

Section

Information Technologies and Computer Engineering

Metrics

Downloads

Download data is not yet available.