Dimensions overall score 4.0

GitHub Code Pulse

No public code linked for this paper yet.

• Instructional alignment, defined as the discrepancy between required and enacted Bloom's taxonomy levels, predicts success in cybersecurity tabletop exercises.result 85%
• Text embeddings and log features outperform models using only Bloom's taxonomy coding for predicting milestone achievement in cybersecurity exercises.result 90%
• A combination of text embeddings and log features provides the best predictive performance for exercise success, achieving a test AUC of approximately 0.80.result 90%
• Bloom's taxonomy frequencies add little predictive value when combined with text embeddings and log features.result 85%
• The open-source INJECT Exercise Platform (IXP) reduces instructor workload and supports the full exercise lifecycle.technical 80%
• Multimodal traces (text and log data) best forecast performance in simulation-based training, while alignment analytics provide interpretable diagnostic insight.result 88%
• A pipeline using predictive models to flag at-risk moments and alignment analytics to provide rationales can create pedagogically actionable feedback.method 75%
• The explanatory power of Bloom alignment alone for predicting milestone achievement is limited, as indicated by a low R-squared value.limitation 85%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card