Cardiac arrhythmia, a frequently encountered cardiac disorder characterized by irregular heart rhythms, poses a substantial risk to a patient's wellbeing. This research explored various algorithms for forecasting cardiac arrhythmias derived from ECG datasets available on Kaggle. Employing machine learning techniques, the study achieved significant accuracy rates: Logistic Regression (71.43%), Random Forest Classifier (58.24%), and Decision Tree Classifier (65.93%). Extracting the different characteristics of cardiac electrograms (ECG), focusing on R peak, QRS duration, and T interval, forms the foundation for arrhythmia categorization. The results are employed to transform arrhythmia management, emphasizing accuracy and key features, with Logistic Regression leading at 73.28%, Random Forest Classifier at 76.82%, and Decision Tree Classifier at 61.22% in AUC-ROC scores. This research contributes to advancing datadriven methodologies in cardiac care, enhancing patient outcomes by facilitating early detection and personalized treatment plans.