Introduction: Diabetes, a global chronic ailment, necessitates advanced predictive models.
Methods: This study leverages the Pima Indian dataset, implementing Logistic Regression (LR), Naïve Bayes, Random Forest, and Decision Tree algorithms. Algorithmic performance is assessed using accuracy, precision, recall, and F1-score.
Results: Table 1 data reveal commendable performance. LR led with 83% accuracy, Decision Tree at 82%, Random Forest at 81%, and Naïve Bayes at 80%. Glucose, BMI, age, insulin, skin fold thickness were identified by the Correlation Heatmap (Fig. 1).
Conclusion: This study affirms the effectiveness of four machine learning algorithms, achieving approximately 80% accuracy in predicting type II diabetes. Findings underscore the relevance and applicability in healthcare scenarios, aligning with real-world considerations.
R. Barakeh: None.
