全球教育展望

传统课堂基于班级授课制，存在低效操练，缺乏个性，师生负担重等问题。本研究构建了以核心素养为导向、基于知识图谱的高中生物学智适应学习系统，通过教学资源的智能化重组赋能教育教学，探索人机协同教学新模式，减负增效，实现大规模的因材施教。本研究将生物学新课标中所有知识点和核心素养要求解构形成知识图谱，支持计算机的推理和计算，匹配开发了15305个微课、动画、试题等资源积件并科学标注，基于知识路径矩阵模型、学习者画像和教学策略模型，构建了生物学智适应学习系统。基于人机协同教育理念，探索了传统课堂融合智能技术的嵌入式、诊断补偿式等教学模式。研究成果先后在150多所高中、4万余名学生中进行了6年实践。实证表明，该系统可赋能教学，实现学习过程可视化、学情诊断精准化、学习资源推送个性化等，减少学生无效操练，提升学习效率，为教师备课和个别辅导提供智能支持；同等学习强度下，学生学业水平考试成绩显著提升；研究形成的生物学领域知识图谱构建技术和推荐算法在物理、数学等学科得到广泛推广。

Traditional classroom teaching subjecting to the class-based teaching system shows several issues such as providing inefficient exercises for students, lacking of personalized learning, and overburdening teachers and students. In this study, a core literacy-oriented, knowledge map-based intelligent adaptive learning system of high school biology is constructed. Through intelligent reorganization of teaching resources, the intelligent adaptive learning system empowers teaching and learning, provide new model of Human-computer Collaborative, reduces the burden and enhances efficiency for students, and thereby teaches students in accordance with their aptitude on a large scale. In this study, all the major concepts and core literacy requirements in the new curriculum standards of biology are deconstructed to form a knowledge map to support computer reasoning and computation. A total of 15,305 pieces of teaching resources including micro-lessons, animations, and test items are developed and labeled in a scientific manner. Based on the knowledge path matrix model, learner profiling and teaching strategy model, an intelligent adaptive learning system of biology is established. By the educational idea of Human-computer Collaborative, embedded and diagnostic compensatory teaching models that combine traditional classrooms with intelligent technology are explored. The achievements have been practiced for six years in more than 150 high schools with more than 40,000 students. The results indicate that the system empowers teaching and contributes to the visualization of learning process, precise diagnosis of students' conditions, and personalized planning of learning resources. It also reduces ineffective exercises for students, improves student learning efficiency, and provides teachers with intelligent support for lesson preparation and individual counselling. Under the same intensity of learning, the students' performance in the Academic Proficiency Tests has been significantly improved. The knowledge mapping techniques and recommendation algorithms developed for biology have also been widely used in physics, mathematics, and other subjects.