Software testing is a method for checking and validating an automated system's ability to fulfill the automation's necessary attributes and functionality with the automation. It is an essential part of software development that is vital to ensuring the quality of a product to be released. The need for automated software testing approaches arises as the operating structures become more complex which requires analyzing software systems behavior to discover faults. Many testing activities are expensive and complex, and the automation of software testing is a realistic approach that has been implemented to get around these problems. At the beginning, when the Waterfall project approach was already commonly applied, testing was introduced to validate the program as an end-of-project solution only before it entered the market. Since then, project methodologies have also evolved, integrating the everpopular Agile, DevOps, and others, requiring more versatile and innovative methods. Machine learning (ML) is one of the new approach introduced to use the groundbreaking technology made possible. Machine Learning is established from the study of pattern recognition and computational learning approach. The main principle reason is to make machines learn without being explicitly programmed. This science absorbs tons of complex data and identifies schemes that are predictive. In this paper, review the state-of-the-art ways in which ML is explored for automating and upgrading software testing is set. And include an overview of the use cases of test automation, an advantage in implementing ML automation techniques along with challenges in current automation testing. The aftereffects of this paper plot the ML viewpoint that are most regularly used to automate software-testing exercises, helping analysts to comprehend the ebb and flow condition of research concerning ML applied to software testing. Its strategies have demonstrated to be very valuable for this automation process and there has been a developing enthusiasm for applying machine learning to mechanize different software testing activities

Keywords : Machine learning (ML), Software testing, Challenges, Use cases of ML based test automation