Cross-project defect prediction: a large scale experiment on data vs. domain vs. process

By Thomas Zimmermann, N. Nagappan, Harald C. Gall, Emanuel Giger, Brendan Murphy. ESEC/SIGSOFT FSE, 2009

719 citations recorded by the source corpus.

Subfields: Industrial and Manufacturing Engineering, Information Systems

View Paper: https://www.semanticscholar.org/paper/87bd02c23c96dc602e3fa31890baaa0613bb75b1

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

1. Ahmed E. Hassan
2. Bram Adams
3. Yasutaka Kamei
4. Sunghun Kim
5. David Lo
6. Shane McIntosh
7. Jaechang Nam
8. Lin Tan

Citations in the corpus

• Heterogeneous Defect Prediction (2015, methodology): “Since the performance of CPDP is usually very poor [59], researchers have proposed various techniques to improve CPDP [29, 37, 51, 54].”
• Software Defect Prediction via Convolutional Neural Network (2017, background): “[36] evaluated the performance of cross-project defect prediction on 12 projects and their 622 combinations.”
• A Systematic Literature Review on Fault Prediction Performance in Software Engineering (2012, neutral): “Our context checklist also overlaps with the 40 project characteristics proposed by Zimmermann et al. [[208]] as being relevant to understanding a project sufficiently for cross project model building (it was impractical for us to implement all 40 characteristics as none of our included studies…”
• A systematic review of machine learning techniques for software fault prediction (2015, background): “Hence, total number of 122 studies [1, 2, 20-139] were identified for further processing and analysis.”
• Transfer learning for cross-company software defect prediction (2012, background): “[9] use decision trees to help managers to estimate precision, recall, and accuracy before attempting a prediction across projects.”
• Online Defect Prediction for Imbalanced Data (2015, methodology): “Our basic experiment setup uses ADTree for its simplicity, easy implementation, and good performance from previous work [26, 65, 66].”
• Researcher Bias: The Use of Machine Learning in Software Defect Prediction (2014, background): “Section 5 gives a discussion of the implications of our findings, a brief review of scientific bias in other disciplines and concludes with three specific recommendations as to how bias might be reduced for future research into empirically evaluating defect prediction systems.”
• Transfer defect learning (2013, background): “However, most approaches are evaluated in within-project settings, i.e., a prediction model is built from a part of a project and the model is evaluated with the remainder of the project by 10-fold cross validation [2], [6], [7], [8] and/or random instance splits [5], [9], [10].”
• An empirical study on software defect prediction with a simplified metric set (2014, methodology): “Results: The study has been conducted on 34 releases of 10 open-source projects available at the PROMISE repository.”
• A large-scale empirical study of just-in-time quality assurance (2013, background): “Zimmermann et al. [62] focus on defect prediction from one project to another using seven commercial projects and four open source projects.”
• Deep Learning for Just-in-Time Defect Prediction (2015, background): “propose a cross-project defect prediction approach; they train a model on a source project which is selected considering several factors, and use the model on a given target project [36].”