Heart disease prediction model using random forest classifier

Omondi, Edwin Elisha (2025) Heart disease prediction model using random forest classifier. World Journal of Advanced Research and Reviews, 26 (2). pp. 3468-3490. ISSN 2581-9615

To forecast the risk of heart disease, I have created a random forest classifier model in this work. I trained the algorithm to identify people into two groups: those at low risk (0) and those at high risk (1) of acquiring heart disease. I did this by utilizing a dataset made up of anonymized patient information. The model's remarkable 88.04% accuracy rate shows how well it can differentiate between the two classes. By thoroughly analyzing the model's performance, I produced an extensive classification report that included information on accuracy overall, precision, recall, and F1-score for every class. The precision and recall metrics demonstrate how well the algorithm can identify individuals who are at high risk of heart disease while reducing the number of false positives. Additionally, the F1-score illustrates the harmony To further visualize the model's categorization results, I created a confusion matrix. The matrix provides further verification of the model's performance by showing the number of true negatives, false positives, false negatives, and true positives. To be more precise, the matrix indicates that the model accurately identified 95 people as high risk and 67 people as low risk, with just 10 and 12 misclassifications, respectively. The accuracy and dependability of the random forest classifier in identifying people at risk of heart disease are demonstrated by this comprehensive investigation. Furthermore, an important advancement in customized medicine is the use of machine learning techniques in the healthcare industry. This model supports ongoing initiatives in preventive medicine and healthcare management by utilizing extensive datasets and advanced algorithms. Since the model is capable of accurately predicting the risk of heart disease, early intervention strategies may be put into effect, which will ultimately improve patient outcomes and lessen the pressure on healthcare systems. To sum up, this random forest classification algorithm shows great promise in precisely identifying people who are at risk of heart disease. This approach supports ongoing initiatives in preventative medicine and personalized healthcare management by fusing sophisticated analytics with clinical insights. The present study's results open new avenues for investigation and progress in the field of cardiovascular health, ultimately leading to improved treatment of patients and public health outcomes.