This paper describes the individual process of programming as a specific part of a more general software development process. We discuss the task of programming process modeling and propose a new approach based on hidden Markov models. The model may be used to solve three kinds of problems: the analysis and comparison of processes, making the process more efficient and less exhausting and transferring the process knowledge from experienced developers to students. The latter two problems are solved by detecting the kind of programmer’s activity. We do it by inferring the most likely sequence of process model states based on observed actions the developer performs using the integrated development environment and other relevant applications.
Citation: Timofeev D.A., Samochadin A.V. Modeling of software development processes with hidden Markov models. St. Petersburg State Polytechnical University Journal. Computer Science. Telecommunications and Control Systems, 2017, Vol. 10, No. 4, Pp. 70–77. DOI: 10.18721/JCSTCS.10406