Phishing is a growing cyber crime that involves posing as a trustworthy entity in order to steal sensitive information. Because of the rise in phishing attacks, there is a greater need for effective solutions to detect and prevent these attacks. Machine learning algorithms have been proposed as a highaccuracy solution for detecting phishing attacks. This paper provides an overview of recent research studies on machine learning-based phishing detection. The survey summarises the methods employed, the evaluation metrics, and the outcomes obtained. In this field, various machine learning algorithms are being utilised, including decision trees, random forests, and support vector machines.The models are developed using a large - scale dataset of phishing and anti-phishing URLs before being applied to new URLs should be classified as phishing or non-phishing..The URL frame, website content, and website reputation can all be used for training. Machine learning offers a promising approach for detecting phishing attacks and has the potential to significantly improve internet users' security. However, more research is needed to improve these systems' performance and robustness against evolving phishing threats. Veredas Journal || ISSN:0874-5102 www.veredasjournal.org Volume 12 Issue 8 2022 Page No: 3