Modeling Hint-Taking Behavior and Knowledge State of Students with Multi-Task Learning
Modeling Hint-Taking Behavior and Knowledge State of Students with Multi-Task Learning
复制标题
DOI：
10.29007/dj6b
复制
发表时间：
2018-07
期刊：
影响因子：
--
通讯作者：
Harvineet Singh;S. Saini;Ritwick Chaudhry;Pradeep Dogga
中科院分区：
文献类型：
--
作者：
Harvineet Singh;S. Saini;Ritwick Chaudhry;Pradeep Dogga
交互式学习环境通过提供提示来填补对概念理解的空白，从而促进学习。研究表明，学习者并未优化利用提示，要么不必要地使用，要么根本不使用。有研究表明，在需要时提供提示可以提高学习成果。学习环境可以使用有效的提示获取预测模型，就是否保留提示或主动提供提示做出适应性决策。以往关于学生行为建模的工作广泛集中在对学习者随时间变化的知识状态进行建模的任务上，这被称为知识追踪。学生行为的其他方面，例如使用提示的倾向，受到的关注有限。以往的知识追踪模型要么忽略使用了提示的问题，要么将使用提示标记为错误回答。我们提出一种多任务记忆增强深度学习模型，以联合预测获取提示的倾向和知识追踪任务。该模型纳入了过去的回答以及在这两个任务中使用提示的影响。我们将该模型应用于两个数据集——2009 - 2010年ASSISTments技能构建数据集和均一教育平台数学练习日志。结果表明，深度学习模型有效地利用了学生回答中存在的序列信息。所提出的模型在提示预测方面比以往的工作显著提高了至少12个百分点。此外，我们证明联合对这两个任务进行建模可提高这两个任务的性能。
Interactive learning environments facilitate learning by providing hints to fill gaps in the understanding of a concept. Studies suggest that hints are not used optimally by learners. Either they are used unnecessarily or not used at all. It has been shown that learning outcomes can be improved by providing hints when needed. An effective hint-taking prediction model can be used by learning environments to make adaptive decisions on whether to withhold or to pro-actively provide hints. Past work on student behavior modeling has focused extensively on the task of modeling a learner's state of knowledge over time, referred to as knowledge tracing. Other aspects of student behavior such as tendency to use hints has garnered limited attention. Past knowledge tracing models either ignore the questions where hints were taken or label hints taken as an incorrect response. We propose a multi-task memory-augmented deep learning model to jointly predict propensity of taking a hint and the knowledge tracing task. The model incorporates the effect of past responses as well as hints taken on both the tasks. We apply the model on two datasets -- ASSISTments 2009-10 skill builder dataset and Junyi Academy Math Practicing Log. The results show that deep learning models efficiently leverage the sequential information present in student responses. The proposed model significantly out-performs the past work on hint prediction by at least 12% points. Moreover, we demonstrate that jointly modeling the two tasks improves performance on both of these.