Globally, heart disease is the main cause of death which affects individuals, irrespective of age and sex. Heart attack, a life-threatening heart disease occurs mainly due to plaque rupture. If it can be diagnosed early, the chance of survival for the patient with a heart disease would be higher. This is where Machine Learning comes in handy. To achieve this goal, this work would detect the chance of heart attack by implementing the best Machine Learning Algorithm. The different ML techniques covered here are XGBoost, CatBoost, GaussianNB, Ridge Regression, etc.…which fall under the category of Supervised learning algorithms. Now, there are several key factors or attributes like Accuracy, Precision, Recall, Execution Time, etc. depending on which the best ML algorithms can be found. To enhance the scores of these key factors Data Manipulation and Data Blending must be done beforehand to remove the missing values from the datasheet. The Technique for Order of Preference by Similarity to the “Ideal Solution” or TOPSIS method under the “multi-criteria decision analysis” (MCDA) was applied in this case, which takes all the necessary factors into account and gives us the best optimal solution (algorithm). The “Ridge Regression” Algorithm achieved a Performance score of 0.96 and has been ranked the best for the early detection of a heart attack.