Dealing with noise in defect prediction

By Sunghun Kim, Hongyu Zhang, Rongxin Wu, Liang Gong. International Conference on Software Engineering, 2011

371 citations recorded by the source corpus.

Subfields: Information Systems, Software

View Paper: https://www.semanticscholar.org/paper/9eaae704216b322f5a17400377fce443853d7b86

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Software

Citations in the corpus

Automated Parameter Optimization of Classification Techniques for Defect Prediction Models (2016, background): “[26, 61], 5 are provided by D’Ambros et al.”
ReLink: recovering links between bugs and changes (2011, methodology): “Kim et al. found that the change classification algorithm [16] is also sensitive to the biased data [18] when the number of instances is small.”
It's not a bug, it's a feature: How misclassification impacts bug prediction (2013, background): “[36] provided guidelines for acceptable noise levels and proposed a noise detection and elimination algorithm.”
The Impact of Class Rebalancing Techniques on the Performance and Interpretation of Defect Prediction Models (2018, neutral): “[28] and Wu et al.”
Transfer defect learning (2013, methodology): “We use these outcomes to evaluate the accuracy of defect prediction with the measures defined below [21], [23], [41]: Precision: The ratio of the number of instances correctly classified as buggy (Nb→b) to the number of instances classified as buggy.”
The impact of tangled code changes (2013, background): “[24] showed in an empirical study that the defect prediction performance decreases significantly when the data set contains 20%-35% of both false positives and false negatives noises.”
An Empirical Comparison of Model Validation Techniques for Defect Prediction Models (2017, background): “The conclusions that are derived from defect prediction models may not be sound if the estimated performance is unrealistic or unstable.”
CoCoNuT: combining context-aware neural translation models using ensemble for program repair (2020, background): “This is a reasonable amount of noise (7%) for ML training on large training data [34, 85].”
The Impact of Automated Parameter Optimization on Defect Prediction Models (2018, background): “[71] find that the randomly-generated noise has a large negative impact on the performance of defect models.”