Using data-mining to enable early interventions in introductory engineering courses

该项目的目标是开发基于技术的技术，直接捕捉和分析学生的学习步骤和途径，然后提供基于该分析的干预措施，以促进工程课程的成功。在本研究中，学生将使用智能笔和平板电脑进行本科工程课程的学习活动。智能笔记录学生解决问题的过程，而平板电脑记录学生如何使用电子书等教学材料。 这种技术组合提供了传统评估方法无法提供的学习过程的细粒度视图，从而为个性化评估提供了强大的工具，可以指导教学。数据挖掘技术被用来检查这些学习活动和考试成绩衡量的学业成绩之间的相关性，提供了一种手段来区分无效和有效的学习策略。这些评估技术被用来建立一个早期预警系统，以确定有学习成绩不佳风险的学生，并建议适当的学习策略。新兴的教育信息学领域已经开始产生关于学生学习的重要见解。然而，这一领域的大多数研究都依赖于来自人工学习环境（如在线辅导系统）的数据。这个项目有助于通过创建技术来捕捉学生在现有工程课程中的学习活动，以适合数据挖掘的形式来改变这一领域。 该项目有助于检测学习过程，并为西班牙裔服务机构的不同学生提出个性化干预措施，从而有助于为传统上代表性不足的群体的学生创造更多的机会，以在STEM职业中取得成功。

The goal of this project is to develop technology-based techniques to directly capture and analyze student learning steps and pathways, and then provide interventions based on that analysis to promote success in engineering courses. In this research, students will use smartpens and tablet computers to carry out learning activities in undergraduate engineering courses. The smartpens record problem-solving work, while the tablets record how students use instructional materials, such as e-books. This combination of technologies provides a fine-grained view of the learning process not available with conventional assessment methods, thereby enabling powerful tools for individualized assessment that can guide instruction. Data mining techniques are used to examine the correlation between these learning activities and academic achievement as measured by exam performance, providing a means to distinguish between ineffective and effective learning strategies. These assessment techniques are used to create an early warning system that identifies students at risk of poor academic performance and recommends suitable learning strategies. The burgeoning field of educational informatics has begun to produce important insights about student learning. However, most research in this field has relied on data from artificial learning environments such as online tutoring systems. This project helps transform this field by creating techniques for capturing students' learning activities in existing engineering courses in a form suitable for data mining. This project helps detect the learning processes and suggest individualized interventions for the diverse students at a Hispanic Serving Institution, and thus helps to create increased opportunities for students from traditionally underrepresented groups to succeed in STEM careers.