Page 64 - FULL REPORT 30012024

Evaluation of the models was done using accuracy measures such as accuracy

score, precision score, recall score, and F1 score. The RF model achieved the

highest accuracy with an F1 score of 96%. The Decision Tree model had an
F1 score of 94%, the Voting Classifier had an F1 score of 91%, and Logistic

Regression performed poorly.

Comparing their results with previous studies, the RF model outperformed

other algorithms in terms of accuracy. The researchers suggested further
improvements by using larger datasets and exploring other machine learning

models like AdaBoost, SVM, and Bagging.

In conclusion, the study demonstrates the effectiveness of machine learning

algorithms in stroke prediction. The RF model showed the highest accuracy in
detecting strokes based on physiological variables. The findings have

implications for early detection and treatment of strokes, potentially benefiting
patients by providing timely medical intervention. Table 2.2 shows the

comparison model’s performance between the previous studies and the current
study.

Table 2.2 Comparative Analysis of Model Performance.

Source: Tazin et al., 2022

47