Data-driven Personalized Learning:Real Problems,Ought-to-be Logic and Implementation Paths
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.
万方数据
维普
