Extend dropout prediction with deep learning, behavioral data, and real-time monitoring¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The authors' future-research agenda specifies four extensions: recurrent neural networks to extract temporal patterns in student behavior; incorporation of behavioral data such as class participation and extracurricular engagement; deployment of a real-time monitoring system to proactively support at-risk students; and adding unit-level performance features such as assignment types, grades received, and class participation for deeper performance insight.

Design Implications¶

Context¶

Requirements¶

• Access to behavioral and unit-level performance data (assignments, grades, participation)
• Infrastructure for real-time monitoring deployment

Constraints¶

Target Learners¶

• higher education students

Target Learning Goals¶

• improved accuracy and timeliness of at-risk student prediction

Related Strategies¶

• Deploy ML-based early-warning prediction to shift retention support from reactive to proactive intervention

Examples¶

-

Key Sources¶

• Gheisari, S., & Salarian, H. (2024). Analysis and Prediction of At-Risk Students using Machine Learning Algorithms. https://scholar.google.com/scholar?q=Analysis+and+Prediction+of+At-Risk+Students+using+Machine+Learning+Algorithms