Modeling Skill Acquisition Over Time with Sequence and Topic Modeling
Modeling Skill Acquisition Over Time with Sequence and Topic Modeling
复制标题
通过序列和主题建模随着时间的推移对技能获取进行建模
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
José P. González
中科院分区：
文献类型：
--
作者：
José P. González
在线教育提供的数据来自学生解决问题的不同水平的熟练程度随着时间的推移。不幸的是，使用这些数据来推断学生知识的方法依赖于昂贵的领域专业知识。我们提出了三种新的数据驱动方法，将序列模型与主题模型联系起来，以推断学生的时变知识。这些方法在复杂性、可解释性、准确性和人类监督方面存在差异。例如，我们最具解释性的方法与领域专家创建的模型具有相似的分类准确性，但需要的工作量要少得多。另一方面，最准确的方法是完全数据驱动的，并且在AUC（分类器的评估指标）中将预测提高了高达15%。
Online education provides data from stu- dents solving problems at different levels of proficiency over time. Unfortunately, meth- ods that use these data for inferring student knowledge rely on costly domain expertise. We propose three novel data-driven meth- ods that bridge sequence modeling with topic models to infer students’ time varying knowl- edge. These methods differ in complexity, interpretability, accuracy and human super- vision. For example, our most interpretable method has similar classification accuracy to the models created by domain experts, but requires much less effort. On the other hand, the most accurate method is completely data- driven and improves predictions by up to 15% in AUC, an evaluation metric for classifiers.