Process Mining Techniques for Collusion Detection in Online Exams

MCML Authors

Andrea Maldonado

→ Group Thomas Seidl
 Database Systems, Data Mining and AI

Thomas Seidl

Prof. Dr.

Director

Database Systems, Data Mining and AI

Abstract

Honesty and fairness are essential. As many skills, practicing those values starts in the classroom. Whether students are examined online or on-site, only testing their knowledge righteously, educators can assess their skills and room for improvement. As online exams increase, we are provided with more suitable data for analysis. Process mining methods as anomaly detection and trace clustering techniques have been used to identify dishonest behavior in other fields, as e.g. fraud detection. In this paper, we investigate collusion detection in online exams as a process mining task. We explore trace ordering for anomaly detection (TOAD) as well as hierarchical agglomerative trace clustering (HATC). Promising preliminary results exemplify, how process mining techniques empower teachers in their decision making, while via flexible configuration of parameters, leaves the last word to them.

inproceedings

---

EduPM @ICPM 2023

2nd International Workshop on Education meets Process Mining at the 5th International Conference on Process Mining. Rome, Italy, Oct 23-27, 2023.

---

Authors

A. Maldonado • L. Zellner • S. Strickroth • T. Seidl

---

Links

DOI

---

Research Area

 A3 | Computational Models