An Initial Exploration of Machine Learning Techniques to Classify Source Code Comments in Real-time

Abstract

Dr. Mohammadi-Aragh investigates the use of digital systems to measure and support engineering education, specifically through learning analytics and the pedagogical uses of digital systems. She also investigates fundamental questions critical to improving undergraduate engineering degree pathways.. She earned her Ph.D. in Engineering Education from Virginia Tech. In 2013, Dr. Mohammadi-Aragh was honored as a promising new engineering education researcher when she was selected as an ASEE Educational Research and Methods Division Apprentice Faculty.

Investigating Novice Developers’ Code Commenting Trends Using Machine Learning Techniques

“…Mohammadi-Aragh et al [9] also assessed the commenting habits of students and categorized them into different types. Beck et al [10] collected student source code comments and labeled them as "sufficient" or "insufficient" according to their codebook from their previous research work and then implemented supervised machine learning techniques. Their results suggest that introducing the lemmatization technique improved the performance of the Random Forest classifier.…”

“…and textual data (e.g., comment text) were used as predictors, both Random Forest and Decision Tree achieved good performance. P. Beck et al [10] evaluated and analyzed the code comments with a single label as either sufficient or insufficient using a binary classifier. In their research study, they only considered the text feature of the comments.…”

“…The main contributions of our study are: This research study has provided the key contributions to taxonomy introduced in [9,10] by analyzing student code comments. Further, we implemented machine learning models to achieve the automated classification of students'/developers' source code comments.…”

“…Building Taxonomy and Dataset Annotation: The generated dataset was carefully analyzed, the annotation for the dataset was performed from the existing taxonomy [9,10], and new categories in the taxonomy were also introduced. Table 4 below contains all the information about the taxonomy, which consists of the comment type name and its description (the types in bold text are newly introduced types in the taxonomy).…”

Herramientas usadas para la evaluación formativa automatizada en cursos de programación asistidos por computadora

“…Las herramientas anteriormente mencionadas se han implementado con diferentes técnicas computacionales, así como también existen herramientas que han incorporado la inteligencia artificial (AI) para la evaluación automatizada en cursos de programación (Srikant and Aggarwal, 2014;Beck et al, 2019;Akram et al, 2020;Dominguez et al, 2010).…”