Використання моделей штучного інтелекту для класифікації числових медичних даних

Abstract

This paper explores the use of artificial intelligence models for the classification of numerical medical data. Machine learning methods are widely used in medical diagnostics to analyze clinical indicators and support decision making. In this study, several models are compared, including Logistic Regression, Random Forest, Gradient Boosting, and Support Vector Machine. The models are evaluated using Accuracy, F1-score, ROC-AUC, and MCC metrics. The experimental results show that ensemble methods provide high classification performance. The Random Forest model demonstrates the best results among the considered algorithms. Based on the obtained results, an approach to improving classification using ensemble models and feature importance analysis is proposed.