SHF: Small: Computer-Aided Grading, Feedback, and Assignment Creating in Massive Online Programming Courses

大规模在线开放课程（mooc）被广泛认为是一项革命性的创新；然而，传统的教学方法并不适用于拥有数万名学生的mooc。该项目旨在设计一套算法工具，在教授计算机编程的mooc课程中，部分自动化三个最基本的教学过程——给作业评分、给学生个性化反馈和创建新作业。这些工具有可能显著提高教师在这些课程中的工作效率，并为学生提供比目前更有效的教育体验。它们还可以指导未来MOOC在编程方面的发展，并在使MOOC模式充分发挥潜力方面发挥作用。在这个项目中开发的算法借鉴了几个不同计算领域的思想。他们利用软件自动推理方面的先进技术，比如自动发现学生代码中的错误并提出修复建议，利用统计学习技术挖掘以前完成的作业的数据库，并推断出一个班级的总体统计数据。最后，本研究还增加了人工计算的维度，例如将同行评价获得的数据植入逻辑和统计分析技术。这些方法适用于任何编程课程；然而，调查人员将通过将他们部署在他们教授的Python编程的特定mooc中来评估他们。

Massive Open Online Courses (MOOCs) are widely regarded as arevolutionary innovation; however, traditional instructionaltechniques do not scale to MOOCs with tens of thousands of students.This project aims to design a set of algorithmic tools to partiallyautomate three of the most basic instructional processes - gradingassignments, giving students personalized feedback, and creating newassignments - in MOOCs that teach computer programming. Such toolshave the potential to significantly raise the productivity ofinstructors in these courses, and give students a far more effectiveeducational experience than what they currently receive. They can alsoguide the development of future MOOCs on programming, and play a rolein making the MOOC model reach its full potential.The algorithms developed in this project draw on ideas from severaldifferent areas of computing. They leverage advances in automatedreasoning about software like automatically finding bugs instudent code and suggesting fixes, and exploit statistical learningtechniques that mine databases of previously-completed assignments andinfer aggregate statistics about a class. Finally, the research has adimension of human computation, for instance seeding logical andstatistical analysis techniques with data obtained through peerevaluation. These methods apply to any programming course; however,the investigators will evaluate them by deploying them in a specificMOOC on Python programming that they teach.