Minimum Information Entropy Based Q-matrix Learning in DINA Model
Minimum Information Entropy Based Q-matrix Learning in DINA Model
复制标题
DINA模型中基于最小信息熵的Q矩阵学习
DOI：
10.1145/2723576.2723653
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Yi Sun
中科院分区：
文献类型：
--
作者：
Shiwei Ye;Yuan Sun;Yi Sun
认知诊断模型 (CDM) 在测试开发和学习者表现测量方面越来越受到关注。 DINA（确定性输入、噪声和门）模型是 CDM 中使用最广泛的模型之一。在本文中，我们提出了一种新方法，并提出了一种交替递归算法来学习 Q 矩阵和不确定性变量、滑动和猜测参数，分别基于 DINA 模型的布尔矩阵分解（BMF）和最小信息熵（MIE）。仿真结果表明，我们的Q矩阵学习算法能够快速收敛到Q矩阵和学生知识状态A矩阵的局部最优解。当该方法扩展到大数据时，这一点尤其重要和适用。
Cognitive diagnosis models (CDMs) are of growing interest in test development and measurement of learners' performance. The DINA (deterministic input, noisy, and gate) model is one of the most widely used models in CDM. In this paper, we propose a new method and present an alternating recursive algorithm to learnQ-matrix and uncertainty variables, slip and guessing parameters, based on Boolean Matrix Factorization (BMF) and Minimized Information Entropy (MIE) respectively for the DINA model. Simulation results show that our algorithm forQ-matrix learning has fast convergence to the local optimal solutions forQ-matrix and students' knowledge statesAmatrix. This is especially important and applicable when the method is extended to big data.