模式识别课程的案例教学改革与实践研究

Case-Based Teaching Reform and Practice Research in Pattern Recognition Course

摘要: 模式识别是人工智能感知与理解外部世界的核心技术，是培养计算机与人工智能人才的关键课程。然而，其教学长期面临内容理论性强、工程实践薄弱、学用脱节的矛盾。传统侧重公式推导的教学模式，导致学生难以将知识迁移至复杂实际问题，学习动力与创新能力不足。为破解此困境，本研究将案例教学法系统引入课程改革，构建了与理论教学深度耦合、分层递进的梯度化案例教学体系，并配套“线上–线下”混合式教学模式与多元化综合评价机制。教学实践证明，该方法能有效激发学习动机，深化理论理解，并显著提升学生的工程实践与创新思维能力。本成果为模式识别及相关工科理论课程的教学改革提供了可行路径与参考。

Abstract: Pattern recognition is a core technology for artificial intelligence to perceive and understand the external world and a key course for cultivating computer and artificial intelligence talents. However, its teaching has long faced the contradiction of strong theoretical content, weak engineering practice, and the disconnection between learning and application. The traditional teaching mode that emphasizes formula derivation makes it difficult for students to transfer knowledge to complex practical problems, resulting in insufficient learning motivation and innovation ability. To address this dilemma, this study systematically introduces the case-teaching method into the curriculum reform, constructs a gradient case-teaching system that is deeply coupled with theoretical teaching and progresses in a hierarchical manner, and is accompanied by an “online-offline” hybrid teaching mode and a diversified comprehensive evaluation mechanism. Teaching practice has proven that this method can effectively stimulate learning motivation, deepen theoretical understanding, and significantly improve students’ engineering practice and innovative thinking abilities. This achievement provides a feasible path and reference for the teaching reform of pattern recognition and related engineering theoretical courses.