Arrhythmia is a prevalent cardiac disorder characterized by irregular heart rhythms, and it necessitates accurate and timely diagnosis for effective treatment. This study utilizes machine learning (ML) techniques to address arrhythmia detection, specifically focusing on the well-established UCI Arrhythmia Dataset. The research entails a comprehensive examination of the dataset, encompassing in-depth analysis, and delves into the various pre-processing steps, feature engineering methods, and the selection of appropriate ML algorithms for arrhythmia classification. The key goal of this study is to assess the performance of multiple ML models and pinpoint the most effective approach for detecting this specific disease within the UCI dataset. The outcomes of this investigation yield valuable insights into the potential of ML in enhancing the accuracy of arrhythmia diagnosis. Consequently, the 88% accuracy with 100% precision has been achieved in the detection of arrhythmia, underscoring its promising potential in clinical applications.