Cost-adaptive multi-level semantic feature learning for source code based bug severity prediction

Puntos clave

• Prediction accuracy for bug severity significantly improves with cost-adaptive learning frameworks.
• A notable improvement of 25% in prediction precision is achieved using multi-level semantic features.
• Cost-adaptive multi-level learning methods leverage specific semantic features from the source code effectively.
• These findings indicate enhanced software quality assurance through better model performance in bug severity assessments.