Humans have an inherent need for healthcare, hence providing it is a mandatory life responsibility. Heart and circulatory system diseases are often referred to as cardiovascular disease. High-risk patients were able to lower their risks thanks to decisions informed by early methods of predicting cardiovascular diseases. Due to the sheer volume of data available in the health care industry, machine learning algorithms are essential for accurately predicting cardiac conditions. The latest studies have focused on finding ways to combine these techniques to create hybrid machine learning algorithms. The proposed research would benefit greatly from the pre-processing of data via measures such as the removal of noise, the removal of missing data, the substitution of default values where appropriate, and the classification of attributes for use in prediction and decision making. The diagnostic model's performance is evaluated using techniques including classification, accuracy, sensitivity, and specificity analysis. This study introduces a methodology for predicting whether a person has heart disease. Accurate models for predicting cardiovascular disease are presented by contrasting the outputs of a number of statistical techniques, including the Support Vector Machine, the Gradient Boosting model, the Random Forest, the Naive Bayes classifier, and the logistic regression technique, on data collected from a given region. To better predict cardiovascular disease, we offer a narrative approach that employs machine learning techniques to identify salient aspects. Different features and multiple classification techniques are proposed to create the prediction model. Through the use of a hybrid random forest and a linear model, we are able to improve the performance of a heart disease prediction model from 80% to 92%.

Cardiovascular Disease Prediction, Machine Learning Techniques, Random forest linear model.