Can I read Hyperparameter Tuning for Software Defect Prediction on EtoBox?

Hyperparameter Tuning for Software Defect Prediction by thi is a document available to read on EtoBox.

What is Hyperparameter Tuning for Software Defect Prediction about?

The study investigates the effectiveness of hyperparameter tuning using Random Search on tree-based classification algorithms for predicting software defects. Utilizing ReLink datasets, the research demonstrates that Decision Tree, Random Forest, and Deep Forest achieved average AUC values of 0.73, 0.79, and 0.79, respectively, with Random Search outperforming other methods. The findings emphasize the advantages of Random Search in optimizing algorithm parameters for improved software defect prediction accu

Author

thi

Language

EN