Generation of Personalized Abstract and Real-World Programming Exercises

Abstract

We present an approach for automatically generating abstract and real-world coding exercises with adjustable difficulty. Using our approach, we can produce exercises based on abstract computational operations which are useful for learning the syntax of a programming language, as well as exercises based on real-life contexts which are useful for learning abstraction and logic formulation. We present the details of our approach as well as an initial evaluation of the exercises by students and teachers of programming. This work can pave the way for the development of intelligent programming tutors with adaptive and personalized feedback that can display content based on the state of the student.