Scholarlink

To solve the problem of insufficient personalization and interpretability of the MOOC platform recommendation system, the user behavior data and course characteristics were analyzed, and an interpretable course recommendation algorithm CountERText based on causal inference, was proposed. This algorithm uses counterfactual reasoning to generate intuitive recommendation explanations, enhances users’ understanding of recommendation logic, optimizes the learning experience, and is of great significance in promoting the intelligent transformation of the education field.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design and Implementation of an Explainable Course Recommendation Algorithm Based on Causal Inference

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xinpeng Chen,
  • Zhengzhou Zhu,
  • Yizhen Xie

摘要

To solve the problem of insufficient personalization and interpretability of the MOOC platform recommendation system, the user behavior data and course characteristics were analyzed, and an interpretable course recommendation algorithm CountERText based on causal inference, was proposed. This algorithm uses counterfactual reasoning to generate intuitive recommendation explanations, enhances users’ understanding of recommendation logic, optimizes the learning experience, and is of great significance in promoting the intelligent transformation of the education field.