Lie detection plays a vital role in various fields, including law enforcement, security and forensic psychology. With the advancement of machine learning techniques, there has been a significant shift towards using computational models for lie detection. This research article presents a comprehensive review of lie detection methods using machine learning approaches. It explores the diverse range of machine learning algorithm employed in deception detection, their performance metrics, and the datasets used for training and evaluation. Additionally, the article discusses the challenges, limitations and future perspectives in the field of machine learning based lie detection. The aim is to provide researchers, practitioners and policymakers with an in-depth understanding of the latest and greatest advancement in this rapidly evolving field.