Educational Reform Practices for Machine Learning Courses in an Interdisciplinary Context

Abstract

The rapid advancement of artificial intelligence has created an unprecedented demand for interdisciplinary talent across various fields. However, teaching machine learning to non-computer science students presents unique challenges. This paper presents innovative teaching approaches and reforms for machine learning education under an interdisciplinary context. The methodology encompasses three main components: foundational knowledge enhancement, interactive case-based teaching, and ChatGPT-assisted learning. The course structure follows an "easy-to-understand basics, progressive learning, and application-oriented" principle, divided into mathematical foundations, programming basics, and practical applications. Through case studies and real-world projects like snack price prediction and traffic flow analysis, students engage in hands-on learning experiences. Additionally, the integration of ChatGPT as a learning tool helps students understand code, debug programs, and optimize machine learning models. This comprehensive teaching model effectively combines theoretical knowledge with practical applications, fostering students' interdisciplinary thinking, programming capabilities, and problem-solving skills.