Analyze learning patterns in a programming assignment
“Disruptive innovation is already at work in higher education, and universities have to look no further than online classes to see examples of change at scale” said Clay Christensen, a Harvard Business School professor. One of the key components of online classes is the Assessment. All of the major learning management systems have some sort of online assessment component and analytics built on top of it. There is not much progress made on analytics for programming assignments. These online platforms are collecting lot of data about how a learner is solving programming assignments, but there is not much feedback provided neither to the learner nor to the instructor based on that data. So focus of this project is to analyze programming assignment data per problem and highlight useful patterns with some visualization both for instructors and learners.
Visualization for Instructors:
Š What are the different ways to solve a problem?
Š Where is my class struggling in solving the problem?
Š Learner activity per problem
Visualization for Learners:
Š Learner progress per assignment
Š Solving pattern per problem
Given a new solution; the above visualizations should be updated in close to real-time.
Some Technologies you will learn/use:
Š Web framework: Ruby on Rails, MEAN stack, or any other web framework
Š Data Analytics: text based clustering
Š Visualization: Use of D3 or similar visualization library.
Š Microservices Architecture
Š Test Driven Development
Š Amazon EC2, MySQL or Amazon DynamoDB, Amazon Elasticsearch