《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (8): 2351-2356.DOI: 10.11772/j.issn.1001-9081.2023081205

• 人工智能 • 上一篇    

基于认知诊断的个性化习题推荐

韩祎珂1, 徐彬1(), 张硕2   

  1. 1.东北大学 计算机科学与工程学院,沈阳 110169
    2.东北大学 信息科学与工程学院,沈阳 110169
  • 收稿日期:2023-09-05 修回日期:2023-12-03 接受日期:2023-12-05 发布日期:2024-08-22 出版日期:2024-08-10
  • 通讯作者: 徐彬
  • 作者简介:韩祎珂(1999—),女,河南新密人,硕士研究生,主要研究方向:故障诊断、智慧教育
    徐彬(1980—),男,浙江兰溪人,副教授,博士,CCF会员,主要研究方向:人工智能、智慧教育、工业智能、故障诊断 xubin@mail.neu.edu.cn
    张硕(1988—),男,辽宁沈阳人,博士研究生,主要研究方向:异常检测、机器学习、需求预测。
  • 基金资助:
    国家自然科学基金资助项目(72271048);辽宁省自然科学基金资助项目(2022?MS?119);中国高校产学研创新基金-云中大学项目(2022MU017)

Personalized exercise recommendation based on cognitive diagnosis

Yike HAN1, Bin XU1(), Shuo ZHANG2   

  1. 1.School of Computer Science and Technology,Northeastern University,Shenyang Liaoning 110169,China
    2.College of Information Science and Technology,Northeastern University,Shenyang Liaoning 110169,China
  • Received:2023-09-05 Revised:2023-12-03 Accepted:2023-12-05 Online:2024-08-22 Published:2024-08-10
  • Contact: Bin XU
  • About author:HAN Yike , born in 1999, M. S. candidate. Her researchinterests include fault diagnosis, smart education.
    XU Bin, born in 1980, Ph. D. , associate professor. His researchinterests include artificial intelligence, smart education, industrialintelligence, fault diagnosis.
    ZHANG Shuo , born in 1988, Ph. D. candidate. His researchinterests include anomaly detection, machine learning, demand forecast.

摘要:

针对现有的基于认知诊断的习题推荐建模角度单一以及习题推荐结果不够合理的问题,提出结合认知诊断和深度因子分解机的个性化习题推荐方法。首先,设计一种知识点关系计算方法构建课程知识树,并提出增强Q矩阵准确表示习题所含知识点的关系的概念;其次,提出基于知识点关系和习题表征的认知诊断(NeuralCD-KD)模型,该模型计算增强Q矩阵,利用特征二阶交叉和注意力机制融合习题难度的内外因素,并模拟学生的认知状态。在私有数据集和公开数据集上验证了提出的认知诊断模型的有效性,并且该方法能对学生的认知状态做出合理解释。为了个性化习题推荐,结合诊断模型和深度双线性因子分解机,提出结合认知诊断和深度因子分解机(NKD-DBFM)方法,在私有数据集上验证了所提习题推荐方法的有效性,在曲线下面积(AUC)上相较于最优基线模型神经认知诊断模型(NeuralCDM)提升了3.7个百分点。

关键词: 认知诊断, 习题推荐, 知识点关系, Q矩阵, 习题表征

Abstract:

A personalized exercise recommendation method that combines cognitive diagnosis and deep factorization machine was proposed to address the problems of single modeling angle and unreasonable exercise recommendation results of the existing exercise recommendation based on cognitive diagnosis. Firstly, a new method for calculating the relationship between knowledge points was designed to construct a course knowledge tree, and the concept of enhanced Q matrix to accurately represent the relationship between knowledge points contained in exercises was proposed. Secondly, the Neural Cognitive Diagnosis with Knowledge-based Discernment (NeuralCD-KD) model was proposed to calculate the enhanced Q matrix. In the model, the feature second-order cross and attention mechanism were used to fuse internal and external factors of exercise difficulty, and the students’ cognitive states were simulated. The effectiveness of the proposed cognitive diagnosis model was verified on private and public datasets, and this method was able to give reasonable explanations for students’ cognitive states. To personalize exercise recommendation, a Neural Knowledge-based Cognitive Diagnosis with Deep Bilinear Factorization Machine (NKD-DBFM) method was proposed by combining the diagnostic model with deep bilinear factorization machine, and the effectiveness of this proposed exercise recommendation method was verified on the private dataset. Compared with the optimal baseline model Neural Cognitive Diagnosis Model (NeuralCDM), the proposed method improves the Area Under Curve (AUC) by 3.7 percentage points.

Key words: cognitive diagnosis, exercise recommendation, knowledge point relationship, Q matrix, exercise representation

中图分类号: