A Machine Learning and Digital Twin-Based Assistance System for Computer Science Students' Career Prediction and Activity Recognition

Abstract

The advancement of digital twin (DT), machine (ML), and deep learning (DL) technology has created new opportunities for students' activity recognition and career guidance in the educational sector. DT and ML technology can be used to update and monitor students’ performance in various courses as well as daily life activity data. Existing research on student activity recognition and career suggestions did not create a dataset that included static, location, accelerometer, and academic data. They did not use digital twin technology for student performance monitoring. The existing ML-based works should have investigated various academic course data and job descriptions for the student's career recommendation with high accuracy. To deal with these issues, this paper creates an ML, DL, and DT-based assistance system for career recommendation and activity recognition based on student personal activity dat and academic course results. Several ML and DL models were tested for career suggestion prediction, and logistic regression achieved the highest accuracy of 98%. The results exhibit that the CNN-based DL model achieves the highest 96% accuracy for the student's daily activity recognition system. According to the performance comparison results, the proposed logistic regression-based career suggestion prediction system outperforms the existing works by at least 4% in accuracy and 7% in precision. The results also show that the proposed CNN-based activity recognition system achieves at least 5% higher accuracy and 3% higher precision values than previous works.

Received: 30 October 2024 | Revised: 19 December 2024 | Accepted: 22 January 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support this work are available upon reasonable request to the corresponding author.

Author Contribution Statament

Tabassum Ferdous: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Visualization, Project administration. Mahfuzulhoq Chowdhury: Conceptualization, Methodology, Writing - original draft, Writing - review & editing, Supervision, Project administration.