Comparative Analysis of Machine Learning Methods: Decision Tree and Kohonen Maps

Data analysis and prediction play a key role in medical diagnosis and decision making. This paper investigates the effectiveness of two data analysis methods, decision tree and Kohonen maps, in the context of thyroid cancer. The study is based on 383 patients and includes analysis of various factors. The results indicate the potential of both methods in accurately predicting disease risk, patient gender and likelihood of recurrence. The decision tree is characterized by its ease of interpretation, while Kohonen maps have the ability to reveal complex dependencies in the data. In conclusion, the article emphasizes the importance of choosing an appropriate data analysis method to achieve accurate and interpretable results.