New benchmark reveals temporal data predicts student dropout

Researchers have developed a new survival benchmark for predicting student dropout in learning analytics. This benchmark harmonizes dynamic and continuous-time representations, comparing various models like Random Survival Forest and Poisson Piecewise-Exponential. The study found that temporal and behavioral data, rather than static demographics, are the most significant predictors of dropout risk. AI

IMPACT Establishes a new standard for evaluating AI models in educational contexts, emphasizing temporal and behavioral data for dropout prediction.

RANK_REASON The cluster contains an academic paper detailing a new benchmark and findings in the field of learning analytics. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.AI →

• Open University Learning Analytics Dataset
• Poisson Piecewise-Exponential
• Random Survival Forest
• XGBoost AFT

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →