Learning from Open-Source Projects: An Empirical Study on Defect Prediction

He, Zhiyuan; Peters, Fayola; Menzies, Tim; Yang, Ye

doi:10.1109/esem.2013.20

Cited by 98 publications

(88 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data transformation approach was based on adjusting the average values of each metric in the training and testing sets (Watanabe et al 2008). The data selection approaches included a Nearest Neighbors filter (Turhan et al 2009), brute force selection method based on combination of projects (He et al 2012), and selection of similar projects combined with feature subset selection (He et al 2013). Transfer learning methods, on the other side, tried to improve the classification algorithmically (Ma et al 2012;Nam et al 2013).…”

Section: Fig 11mentioning

confidence: 99%

Assessment and cross-product prediction of software product line quality: accounting for reuse across products, over multiple releases

et al. 2014

View full text Add to dashboard Cite

The goals of cross-product reuse in a software product line (SPL) are to mitigate production costs and improve the quality. In addition to reuse across products, due to the evolutionary development process, a SPL also exhibits reuse across releases. In this paper, we empirically explore how the two types of reuse-reuse across products and reuse across releases-affect the quality of a SPL and our ability to accurately predict fault proneness. We measure the quality in terms of post-release faults and consider different levels of reuse across products (i.e., common, high-reuse variation, low-reuse variation, and single-use packages), over multiple releases. Assessment results showed that quality improved for common, low-reuse variation, and single-use packages as they evolved across releases. Surprisingly, within each release, among preexisting ('old') packages, the cross-product reuse did not affect the change and fault proneness. Cross-product predictions based on pre-release data accurately ranked the packages according to their post-release faults and predicted the 20 % most faulty packages. The predictions benefited from data available for other products in the product line, with models producing better results (1) when making predictions on smaller products (consisting mostly of common packages) rather than on larger products and (2) when trained on larger products rather than on smaller products.

show abstract

Section: Fig 11mentioning

confidence: 99%

Assessment and cross-product prediction of software product line quality: accounting for reuse across products, over multiple releases

et al. 2014

View full text Add to dashboard Cite

show abstract

“…However, other criteria can be considered for a comparative analysis. For example, in their original published work [5,12], both 2012Ma and 2013He presented lower computation time cost in relation to the 2009Turhan solution, although 2013He presented higher complexity in relation to 2012Ma. On the other hand, 2013He is more robust to redundant and irrelevant attributes in relation to 2012Ma.…”

Section: Rq1: Which Cpdp Methods Perform Better Across Datasets?mentioning

confidence: 97%

“…At least three different strategies can be identified among the transfer learning solutions: transformation of data [8,9,42,43]; filtering a subset of the training data [10,44,11,12]; and weighting the training data according to the target data [5].…”

Section: Cross-project Defect Predictionmentioning

confidence: 99%

“…Herbold [11] put forward the use of the k-nearest neighbour filter but to select the most similar projects (instead of individual examples). He et al [12] suggested the selection of the most similar projects but considering the separability of each project in comparison to the target data. Moreover, they also suggested the use of the separability strategy to filter unstable attributes.…”

Section: Cross-project Defect Predictionmentioning

confidence: 99%

“…However, to the best of our knowledge, no comprehensive comparison between such transfer learning solution methods has been published to date. Existing comparative analyses are restricted to a few methods [5,12,2] or do not provide a uniform experimental setup [4].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cross-project defect prediction with meta-Learning

Porto¹

View full text Add to dashboard Cite

The prediction of defects in a target project based on data from external projects is called Cross-Project Defect Prediction (CPDP). Several methods have been proposed to improve the predictive performance of CPDP models. However, there is a lack of comparison among state-of-the-art methods. Moreover, previous work has shown that the most suitable method for a project can vary according to the project being predicted. This makes the choice of which method to use difficult. We provide an extensive experimental comparison of 31 CPDP methods derived from state-of-the-art approaches, applied to 47 versions of 15 open source software projects. Four methods stood out as presenting the best performances across datasets. However, the most suitable among these methods still varies according to the project being predicted. Therefore, we propose and evaluate a meta-learning solution designed to automatically select and recommend the most suitable CPDP method for a project. Our results show that the meta-learning solution is able to learn from previous experiences and recommend suitable methods dynamically. When compared to the base methods, however, the proposed solution presented minor difference of performance. These results provide valuable knowledge about the possibilities and limitations of a meta-learning solution applied for CPDP.

show abstract

An improved transfer adaptive boosting approach for mixed‐project defect prediction

Gong

Jiang

2019

J Software Evolu Process

View full text Add to dashboard Cite

Software defect prediction (SDP) has been a very important research topic in software engineering, since it can provide high-quality results when given sufficient historical data of the project.Unfortunately, there are not abundant data to bulid the defect prediction model at the beginning of a project. For this scenario, one possible solution is to use data from other projects in the same company. However, using these data practically would get poor performance because of different distributional characteristics among projects. Also, software has more non-defective instances than defective instances that may cause a significant bias towards defective instances.Considering these two problems, we propose an improved transfer adaptive boosting (ITrAd-aBoost) approach for being given a small number of labeled data in the testing project. In our approach, ITrAdaBoost can not only employ the Matthews correlation coefficient (MCC) as the measure instead of accuracy rate but also use the asymmetric misclassification costs for non-defective and defective instances. Extensive experiments on 18 public projects from four datasets indicate that: (a) our approach significantly outperforms state-of-the-art cross-project defect prediction (CPDP) approaches, and (b) our approach can obtain comparable prediction performances in contrast with within project prediction results. Consequently, the proposed approach can build an effective prediction model with a small number of labeled instances for mixed-project defect prediction (MPDP).

show abstract

Learning from Open-Source Projects: An Empirical Study on Defect Prediction

Cited by 98 publications

References 26 publications

Assessment and cross-product prediction of software product line quality: accounting for reuse across products, over multiple releases

Assessment and cross-product prediction of software product line quality: accounting for reuse across products, over multiple releases

Cross-project defect prediction with meta-Learning

An improved transfer adaptive boosting approach for mixed‐project defect prediction

Contact Info

Product

Resources

About