Investigating the effect of dataset size, metrics sets, and feature selection techniques on software fault prediction problem

By Cagatay Catal, Banu Diri. Information Sciences, 2009

343 citations recorded by the source corpus.

Subfields: Information Systems, Software, Signal Processing

View Paper: https://www.semanticscholar.org/paper/5811f12d3907ccf4de46cfe568be7ad777182c1a

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. Xin Yao
2. Shuo Wang
3. Peng He
4. Xiao Liu
5. Bing Li
6. Yutao Ma
7. Jun Chen
8. Victor Robles

In Software

1. Xin Yao
2. Shuo Wang
3. Peng He
4. Xiao Liu
5. Bing Li
6. Yutao Ma
7. Jun Chen
8. Victor Robles

In Signal Processing

1. Xin Yao
2. Shuo Wang
3. Victor Robles
4. Min-Ling Zhang
5. José María Peña Sánchez
6. Cagatay Catal

Citations in the corpus

• Feature selection for multi-label naive Bayes classification (2009, methodology): “Note that other feature selection techniques may also be used in place of PCA and GA [25,6,36,19] in order to enhance MLNB-BASIC.”
• Distributed Intrusion Detection System in a Multi-Layer Network Architecture of Smart Grids (2011, unknown): “until all the antigens are tested and then the classification will take place by performing the nearest neighbor search in the memory pool [49].”
• Searching for rules to detect defective modules: A subgroup discovery approach (2012, background): “…the False negative rate (FNr ¼ FNTPþFN) is the proportion of positive instances misclassified as belonging to the negative class; the True negative rate (TNr ¼ TNFPþTN) is the proportion of negative instances correctly classified (specificity); and finally, the false positive rate (FPr ¼…”
• Transfer learning for cross-company software defect prediction (2012, background): “But later, Catal and Diri [7] investigate the effects of data set size, metrics set, and feature selection techniques on performance of prediction model, and show that prediction algorithm is more important than the metrics suite.”
• An empirical study on software defect prediction with a simplified metric set (2014, methodology): “Keywords: defect prediction, software metrics, metric set simplification, software quality”
• Using Class Imbalance Learning for Software Defect Prediction (2013, methodology): “Random Forest is shown to be ineffective, as the bootstrapping training strategy and the tree learner are sensitive to class imbalance [53] [9].”
• Software fault prediction: A literature review and current trends (2011, methodology): “Benefits of software fault prediction are listed as follows (Catal & Diri, 2009a): Reaching a highly dependable system, Improving test process by focusing on fault-prone modules, Selection of best design from design alternatives using objectoriented metrics, Identifying refactoring candidates that…”
• RCLinker: Automated Linking of Issue Reports and Commits Leveraging Rich Contextual Information (2015, unknown): “, the number of defects that affect various classes in a project during its life-time), which can, in turn, be used to build effective bug prediction solutions [24], [7], [43], [28], [41], [63], [21].”