SOLVING BINARY CLASSIFICATION PROBLEM USING MACHINE LEARNING METHODS

Alexandra A. Danshina

Student, Department of Information Security,
Volgograd State University
This email address is being protected from spambots. You need JavaScript enabled to view it.
Prosp. Universitetsky, 100, 400062 Volgograd, Russian Federation

Aleksey A. Babenko

Candidate of Sciences (Pedagogy), Associate Professor,

Department of Information Security,
Volgograd State University
This email address is being protected from spambots. You need JavaScript enabled to view it.

Prosp. Universitetsky, 100, 400062 Volgograd, Russian Federation

Abstract. This study discusses machine learning models for solving binary classification problems. An algorithm for processing a data set for training and testing is

provided, as well as a comparative analysis of the proposed models, based on the results of which the most rational one for achieving the stated goal was determined. A structured dataset consisting of 86 records on seven criteria was created, with categorical variables such as location, merchant and gender transformed using one-point coding. Correlation analysis was performed to assess the relationships between the numerical features. The dataset was then divided into training (70%) and test (30%) subsets for model evaluation. The different machine learning models were compared using F1-Score metric. All considered machine learning models cope with the objective of solving the binary classification problem. However, the AdaBoost model, used in conjunction with a weak single-level decision tree algorithm, turned out to be the most rational in use. High efficiency is achieved by bousting weak classifiers, which also compensates for such a problem of the model as overtraining.

Key words: binary classification, machine learning models, data set, data distribution, correlation.

This work is licensed under a Creative Commons Attribution 4.0 International License.

Attachments:

1_Danshina, Babenko.pdf URL: https://ti.jvolsu.com/index.php/en/component/attachments/download/946

1 Download