数据表征学习过程及其应用 ——学习分析数据集国际研究综述

How to Use the Data to Describe the Learning Process ——A Review of Data Sets Used in Learning Analytics

教育信息系统中存储着大量不同格式和不同粒度级别的学生数据，这些数据为利用学习分析技术发现和解决学生学习过程中的问题提供了可能。然而，如何使用恰当的数据表征学生的学习过程，对研究者来说仍存在很大挑战。本文首先从数据来源、具体数据项、数据项收集方法、在学习分析中的应用四个方面对现有学习分析研究中使用的数据集进行了整理和归纳，然后对现有数据集的研究现状、未来发展趋势进行了讨论和分析。以期呈现现有学习分析领域数据集应用的概貌，为后续学习分析中数据集的研究奠定基础。

Huge amount of learners’ dataset is stored in education information systems with different formats and different granularity levels. These data provide the possibility to find and solve the problems of students in the learning process using learning analytics. However， how to use the appropriate data to describe the learning process of learners is still a great challenge for researchers. The data set is one of the core issues of Learning Analytics research. This article firstly provides a review of the typical data sets of learning analytics. Then we analyze and compare the objectives of analysis， the categories of models， the sources of data， the specific data items and the methods of analysis of the typical data sets. Finally， this paper presents the summary and analysis of the current situation and trend of research on data models of learning analytics. This article shows a general picture of data sets in the field of learning analytics and lays the foundation for subsequent research in the future.