计算机科学 ›› 2017, Vol. 44 ›› Issue (10): 265-268.doi: 10.11896/j.issn.1002-137X.2017.10.048

• 人工智能 • 上一篇    下一篇

二值矩阵分解的认知建模方法研究

张猛,付丽华,何婷婷,杨青   

  1. 华中师范大学计算机学院 武汉430079;华中师范大学教育信息化协同创新中心 武汉430079,中国地质大学武汉数学与物理学院 武汉430074,华中师范大学计算机学院 武汉430079,华中师范大学计算机学院 武汉430079
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受教育部新世纪人才计划(NCET-13-1011),华中师范大学中央高校基本科研业务费教育科学专项(CCNU16JYKX21),中央高校科研业务费项目(CCNU15A05022)资助

Cognitive Modeling Based on Binary Matrix Factorization

ZHANG Meng, FU Li-hua, HE Ting-ting and YANG Qing   

  • Online:2018-12-01 Published:2018-12-01

摘要: 根据考试反馈数据,提出新颖的逻辑斯提克二值矩阵分解方法,来预测未来的学生考试成绩并自动对考题进行模式分类,同时设计新的算法对建模中遇到的非凸优化问题进行求解。在模拟数据和真实的美国SAT考试数据上进行的实验发现,新方法不仅可以准确地预测学生的考试表现,而且能够将考题按照知识点进行自动模式分类。实验结果表明, 新的方法相比经典方法在结果的可解释性和估计精度方面有明显的提升。

关键词: 认知建模,二值矩阵分解,考题分类,学生成绩预测

Abstract: A novel logistic binary matrix factorization (LBMF) was proposed to predict the students’ performance and to classify the exam items.Besides a new algorithm was designed to tackle the non-convex optimization problem involved in LBMF.The experiments are performed on both simulated data and real data.The results indicate that LBMF can not only predict the students’ academic performance but also classify the examination items according to the know-ledge points they require.And it can be concluded that LBMF outperforms significantly the out-of-date algorithms in the applications.

Key words: Cognitive modeling,Binary matrix factorization,Item classification,Student performance prediction

[1] LAN A,WATERS A,STUDER C,et al.Sparse Factor Analysis for Learning and Content Analytics[J].Journal of Machine Learning Research,2014,15(6):1959-2008.
[2] WU R,LIU Q,LIU Y,et al.Cognitive Modelling for Predicting Examinee Performance[C]∥Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJCAI 2015).2015:1017-1024.
[3] LAN A,GOLDSTEIN T,BARANIUK R.Dearlbreaker:A Nonlinear Latent Variable Model for Educational Data[C]∥Proceedings of the 33rd International Conference on Machine Lear-ning(ICML 2016).2016:266-275.
[4] RASCH G.Probabilistic Models for Some Intelligence and Attainment Tests [M].MESA Press,1993.
[5] TORRE J.The generalized DINA model framework[J].Psy-chometrika,2011,76(2):179-199.
[6] DESMARAIS M.Mapping Question Items to Skill with Non-negative Matrix Factorization[J].ACM KDD-Explorations,2011,13(2):30-36.
[7] DESMARAIS M,MEHESHTI B,NACEUR R.Item to SkillMapping:Deriving a Conjunctive Q-matrix from Data[C]∥Proceedings of the 11th International Conference Intelligent Tutoring Systems.2012:454-463.
[8] SUN Y,YE S,INOUE S,et al.Alternating Recursive Method ofQ-matrix Learning [C]∥Proceedings of the 7th Educational Data Mining.2014:14-20.
[9] ZHANG Z,LI T,DING C.Binary Matrix Factorization for Analyzing Gene Expression Data[J].Data Mining and Knowledge Discovery,2010,0(1):28-52.
[10] LI T.A General Model for Clustering Binary Data[C]∥Pro-ceedings of the 11th International Conference on Knowledge Discovery and Data Mining.2005:188-197.
[11] SHEN B,JI S,YE J.Mining Discrete Patterns via Binary Matrix Factorization [C]∥Proceedings of 15th International Conference on Knowledge Discovery and Data Mining.2009:757-966.
[12] XU Y,YIN W.A Block Coordinate Descent Method for Regularized Multi-convex Optimization with Applications to Nonne-gative Tensor Factorization and Completion[J].SIAM Journal of Imaging Sciences,2013,6(3):1758-1789.
[13] WINTERS T.Datasets for Topic Clustering [DB/OL].http://alumni.cs.ucr.edu/~titus.
[14] HUANG J M.Application of Bayesian Network to predictingStudents’ Achievement[J].Computer Science,2012,39(11A):280-282.(in Chinese) 黄建明.贝叶斯网络在学生成绩预测中的应用[J].计算机科学,2012,39(11A):280-282.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!