Heart and blood vessel disorders are referred to as cardiovascular diseases (CVDs). It is one of the leading global cause of death and consists of many disorders that harm cardiovascular system. The World Health Organization (WHO) estimates that in 2019, 18 million deaths worldwide were caused by CVDs, accounting for about 32% of all deaths. Therefore, early detection and prediction of cardiovascular disease can be beneficial in identifying high-risk individuals and enabling timely interventions to reduce the disease's impact and improve patient outcomes. This research provides a machine learning (ML)-based framework CVD detection to satisfy this criterion. The proposed model includes data preprocessing, hyperparameter optimization using GridSearchCV, and classification by supervised learning approaches such as support vector machine (SVM), K-nearest neighbors (KNN), XGBoost, random forest (RF), LightBoost (LB), and stochastic gradient descent (SGD). All these models are carried out on the publicly accessed database, namely Kaggle. The experimental results demonstrate that the suggested ML technique has attained 92.76% detection rate with the SGD classifier on the 80:20 training/testing ratios which is superior to the well-received approaches.