It's not a bug, it's a feature: How misclassification impacts bug prediction

By Kim Herzig, Sascha Just, Andreas Zeller. International Conference on Software Engineering, 2013

477 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Information Systems, Software

View Paper: https://www.semanticscholar.org/paper/86a0e7286fb27d8e817c3831982d50cadddc498f

How researchers see this paper

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

In Computer Networks and Communications

In Information Systems

In Software

Citations in the corpus

An Empirical Study on Learning Bug-Fixing Patches in the Wild via Neural Machine Translation (2018, unknown): “As the content of commit messages and issue trackers might imprecisely identify bug-fixing commits [4, 29], two authors independently analyzed a statistically significant sample (95% confidence level ±5% confidence interval, for a total size of 384) of identified commits to check whether they were a”
AR-miner: mining informative reviews for developers from mobile app marketplace (2014, background): “Specifically, the first category of related work in this field is to address problems in bug repositories [38, 9, 8, 25].”
Bug report, feature request, or simply praise? On automatically classifying app reviews (2015, background): “[15] conducted a study and found that about a third of all manually-classified reports in the issue trackers of five large open source projects are misclassified, e.”
Online Defect Prediction for Imbalanced Data (2015, methodology): “Manually veri.ed bug reports are available for Lucene and Jackrabbit [22], which are used in our study for extracting more accurate bug-.xing changes.”
The impact of tangled code changes (2013, methodology): “[27] split each change set into a set of individual change operations that added or deleted method calls or method definitions.”
The Impact of Automated Parameter Optimization on Defect Prediction Models (2018, background): “, noise generated by actually mislabelled issue reports [53]) does not typically impact the precision of defect predictionmodels.”