Integrating K-means clustering into automatic programming assessment tool for student performance analysis

Rina Harimurti, Ekohariadi Ekohariadi, Munoto Munoto, I. G. P. Asto Buditjahjanto

Abstract


Computer programming is a subject involving a large number of logic programming activities. A programmer is compulsory to master skills of algorithms, logic, and programming language to conduct programming. An An Automatic Programming Assessment Tool is an automated tool used to assist instructors in assessing programming tasks. The technology used in this application is Open-Source based with an evaluation module that will evaluate the sent program code, assessment, and classification. The evaluation results were then processed in the assessment module, where a comparison process with the test case was performed along with the point calculation. The classification module was used to divide students into five groups based on the point of each practicum. This study used K-Means Clustering classification method. The entities included were lecturers, assistants, students, and compilers. This application had 2 levels of users namely admin and students. Scoring results were then used in the process of determining the classification of student’s performance based on the K-Means Clustering method. In connection with the classification test results with three iterations, three practicum scores resulted that the classification process was successfully carried out with student’s performance divided into five groups covering very good, good, sufficient, less, and very less. The data used in the clustering process consisted of 41 students with 10 attributes which were then grouped into 3 groups (clusters).

Keywords


Automatic programming assessment; K-means clustering; Programming



DOI: http://doi.org/10.11591/ijeecs.v22.i3.pp%25p

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics