Phishing emails pose a significant threat to individuals and organizations by exploiting trust to obtain sensitive information, financial data, and account credentials. Traditional phishing detection approaches, often rule-based, struggle to keep pace with the evolving tactics used by cyber criminals. Consequently, machine learning (ML) has emerged as a powerful tool in enhancing phishing detection by identifying subtle patterns and anomalies within emails that indicate phishing intent. This survey presents a comprehensive review of recent advances in machine learning techniques applied to phishing email detection. It explores a range of approaches, from supervised learning models such as logistic regression, decision trees, and support vector machines to more complex models like deep learning, neural networks, and natural language processing (NLP). The survey discusses the role of feature engineering in detecting phishing characteristics, including URL analysis, sender information, email content, and embedded links. Furthermore, it addresses the challenges faced in phishing detection, such as dealing with high-dimensional data, reducing false positives, and adapting models to dynamic phishing strategies. This review concludes with insights into current trends, limitations, and future research directions in the field of machine learning-based phishing email detection, underscoring the potential of integrating ML with real-time systems for robust and adaptive email security.