Towards Program Comprehension: Knowledge Profiles for C Programmers

Abstract Following up on the issue of programmer knowledge profiles, the present article deals with the generation of profiles for C language programmers. The main idea is based on static analysis of source codes, assembling the most significant data about their author. Such a profile can point out some abilities and habits of the programmer. The process of static analysis applies methods and techniques of size metrics, complexity metrics, and clone detection. We also present an experiment focused on novice programmers. The results show that knowledge profiles provide an easy way of novice programmers’ progress tracking. The main contribution is believed to be in the area of program comprehension.