| Summary: | AI-enabled Interactive learning environment has become an important platform. It needs to reduce or improve the ineffective and invalid learning behavior, and build a timely and effective diversion inference model of learning effectiveness to provide early warning. Through the early warning, Timely identify risk learners, make effective early warning responses and make targeted interventions. Early warning is an automatic measure based on learning behavior, it evaluates the risk trend, calculates the possible probability of learners' failure in assessment and leaving early. This study proposes and designs a diversion inference model of learning effectiveness. Firstly, we analyze and design one differential evolution strategy. Secondly, the convolutional neural network is improved and optimized based on the differential evolution strategy. Thirdly, the corresponding algorithms are designed and applied. Finally, based on the visualization of the experimental results, we put forward the intervention measures. The technical means might serve the precise teaching intervention and learning recommendation, and has strong theoretical value and practical significance for AI-enabled interactive learning processes.
|
|---|