【作 者】：
钟绍春，杨 澜，范佳荣
【关 键 词 】：
个性化学习； 数据驱动； 情境感知； 学习路径规划； 教育知识图谱
【栏 目】：
理论探讨
【中文摘要】：
教育数字化转型的全面推进和人工智能在教育中的广泛应用，为破解个性化学习难题提供了切实可行的途径，数据驱动的个性化学习已成为教育高质量发展的必由之路。然而，当前数据驱动的个性化学习普遍存在着学习行为感知与状态评价精度不高、学习特征挖掘不准、学习规律挖掘不全、学习问题溯源不深、学习干预精度不佳等瓶颈性难题。为此，研究从情境感知、主体理解和智能干预等方面深入剖析了数据驱动个性化学习的应然逻辑。在此基础上，从学习行为数据有效感知与理解、学习效果精准评估的个性化学习追踪、薄弱知识点和异常学习行为的学习问题成因溯源、潜在交互学习规律发现的教育知识图谱高阶推理、公共学习路网构建与高适配个性化学习路径规划等方面，讨论了数据驱动个性化学习的实现路径和方法。
【英文摘要】：
The comprehensive advancement of educational digital transformation and the widespread application of artificial intelligence in education have provided a practical way to solve the problem of personalized learning, and data-driven personalized learning has become a necessary path for high-quality education development. However, the current data-driven personalized learning is generally characterized by bottleneck problems such as low precision of learning behavior perception and state evaluation, inaccurate mining of learning features, incomplete mining of learning laws, insufficient tracing of learning problems, and poor precision of learning intervention. Therefore, the study analyzes the ought-to-be logic of data-driven personalized learning from the aspects of context perception, subject understanding, and intelligent intervention. Based on this, from the effective perception and understanding of learning behavior data, personalized learning tracking with precise assessment of learning effects, tracing the causes of learning problems of weak knowledge points and abnormal learning behaviors,high-order reasoning of educational knowledge graph for discovering potential interactive learning laws, the construction of public learning networks and the planning of high-adaptive personalized learning paths, the study discusses the implementation path and methods of data-driven personalized learning.