Early Identification of Novice Programmers' Challenges in Coding Using Machine Learning Techniques

Publisher:
ACM
Publication Type:
Conference Proceeding
Citation:
Proceedings of the 2016 ACM Conference on International Computing Education Research, 2016, pp. 263 - 264 (2)
Issue Date:
2016-08-25
Full metadata record
Files in This Item:
Filename Description Size
AA_MS_Submit.pdfAccepted Manuscript version148.36 kB
Adobe PDF
It is well known that many first year undergraduate university students struggle with learning to program. Educational Data Mining (EDM) applies machine learning and statistics to information generated from educational settings. In this PhD project, EDM is used to study first semester novice programmers, using data collected from students as they work on computers to complete their normal weekly laboratory exercises. Analysis of the generated snapshots has shown the potential for early identification of students who later struggle in the course. The aim of this study is to propose a method for early identification of "at risk" students while providing suggestions on how they can improve their coding style. This PhD project is within its final year.
Please use this identifier to cite or link to this item: