This paper presents a comprehensive review of fuzzy machine learning (FML), covering theory, approaches, and applications. FML integrates fuzzy techniques such as fuzzy sets, fuzzy systems, fuzzy logic, fuzzy measures, and fuzzy relations with traditional machine learning to address uncertainties in data. The review categorizes FML into five areas: fuzzy classical machine learning, fuzzy transfer learning, fuzzy data stream learning, fuzzy reinforcement learning, and fuzzy recommender systems. The paper highlights how fuzzy techniques enhance traditional machine learning by handling imprecise data, ambiguous relationships, and noisy information. It discusses the advantages of FML in complex and uncertain environments, including improved interpretability and robustness. The review also identifies key challenges in traditional machine learning and how FML addresses them. The paper provides a detailed analysis of recent developments in FML, including the application of fuzzy techniques in transfer learning, data stream learning, and reinforcement learning. It emphasizes the importance of FML in real-world applications where uncertainty is prevalent, such as robotics, autonomous vehicles, and medical diagnosis. The review concludes with a discussion of the state-of-the-art FML models and future research directions.This paper presents a comprehensive review of fuzzy machine learning (FML), covering theory, approaches, and applications. FML integrates fuzzy techniques such as fuzzy sets, fuzzy systems, fuzzy logic, fuzzy measures, and fuzzy relations with traditional machine learning to address uncertainties in data. The review categorizes FML into five areas: fuzzy classical machine learning, fuzzy transfer learning, fuzzy data stream learning, fuzzy reinforcement learning, and fuzzy recommender systems. The paper highlights how fuzzy techniques enhance traditional machine learning by handling imprecise data, ambiguous relationships, and noisy information. It discusses the advantages of FML in complex and uncertain environments, including improved interpretability and robustness. The review also identifies key challenges in traditional machine learning and how FML addresses them. The paper provides a detailed analysis of recent developments in FML, including the application of fuzzy techniques in transfer learning, data stream learning, and reinforcement learning. It emphasizes the importance of FML in real-world applications where uncertainty is prevalent, such as robotics, autonomous vehicles, and medical diagnosis. The review concludes with a discussion of the state-of-the-art FML models and future research directions.