SCFH: A Student Analysis Model to Identify Students’ Programming Levels in Online Judge Systems

Abstract

Computer basic teaching is an essential basic learning content in higher education teaching. In order to encourage students and enable them to practice and improve their programming ability, the online judge system has been introduced into the programming course for compiling, executing and evaluating the algorithm source code submitted by students. The asymmetry of students’ programming level is an important issue when teachers guide the programming of online judge system. We used the exploratory factor analysis method to identify the potential variable structure from the log data submitted by the students of the online judge system, and evaluate the programming level of the students to predict the “at risk” learners. We proposed a student participation model, SCFH, based on this variable structure. Using the log data of the students in the C language course and their final exam results, we trained a deep neural network based on SCFH to divide the students into three different grades, namely “risky”, “intermediate” and “advanced”. To verify the validity of the model, we used the prediction model to classify students in another C++ language programming course. The results show that the submission log data model SCFH can be used to predict the programming ability of students, and the validity of these results can be tested by examination results.

Prediction of Coding Intricacy in a Software Engineering Team through Machine Learning to Ensure Cooperative Learning and Sustainable Education

“…A deep neural network is trained on the data to classify the students as risky, intermediate, and advanced. This system proved its performance by correlating with the examination result [30].…”

Identifying Student Profiles Within Online Judge Systems Using Explainable Artificial Intelligence

“…In this context in which syntactic-level errors are less common than semantic ones, most approaches rely on the use of OJ systems and ML-based analysis techniques. Examples in the literature include the work by [37] that proposes the use of a supervised classifier to predict final grades based on activity data, that of [38] that studies the correlation between the different features from data related to the assignments of the students and the final grades with linear models, [39] that addresses the problem as an exploratory factor analysis task, or the work by [40] that combined data from an OJ with static information about the students-demographic information or academic marks obtained before enrolling in the course, among others-to predict their performance before each intermediate exam and, accordingly, suggest corrective actions with those who are likely to underperform. Note that, while successful, the main drawback of these proposals is the lack of interpretability of the derived models.…”

Design of an Online Programming Platform and a Study on Learners’ Testing Ability

“…Research by Xu Bin and others also shows that students' programming ability is related to the number of submissions in the online review system. They used an exploratory factory analysis model to identify underlying variable structures from log data submitted by students in an online assessment system and assess students' programming proficiency to predict "at-risk" learners [34]. In our approach, we extend this model to consider the factor of program complexity because complex programs need more tries and submissions.…”