Using K-core Decomposition on Class Dependency Networks to Improve Bug Prediction Model's Practical Performance

Qu, Yu; Zheng, Qinghua; Chi, Jianlei; Jin, Yingze; He, Ancheng; Cui, Di; Zhang, Hengshan; Liu, Ting

doi:10.1109/tse.2019.2892959

Cited by 47 publications

(16 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The quantity

is a small value, controlling the convergence of the membership

. In our experiments,

, that is, our results obtained converged to the 0.01 specification level for comparing the two different methods [ 32 , 33 ]. The simulation lasted for 5000 time steps.…”

Section: Performance Evaluationmentioning

confidence: 62%

An Energy-Efficient Clustering Method for Target Tracking Based on Tracking Anchors in Wireless Sensor Networks

2022

Sensors

View full text Add to dashboard Cite

As a key technology in wireless sensor networks (WSNs), target tracking plays an essential role in many applications. To improve energy efficiency, clustering is widely used in tracking to organize the network to achieve data fusion and reduce communication costs. Many existing studies make dynamic adjustments based on static clusters to track moving targets. However, the additional overhead caused by frequent cluster reconstruction and redundant data transmission is rarely considered. To address this issue, we propose a tracking-anchor-based clustering method (TACM) in this paper, in which tracking anchors are introduced to provide activation indications for sensors according to the target position. We use the rough fuzzy C-means (RFCM) algorithm to locate the anchors and use the membership table to activate sensors to form a cluster. Since there are no sending, receiving, and fusing data tasks for anchors, they are lightly burdened and can significantly reduce the frequency of being rotated. Moreover, the state of cluster members (CMs) is scheduled using the linear 0–1 programming to reduce redundant transmissions. The simulation results demonstrate that, compared with some existing clustering methods, the proposed TACM effectively reduces the energy consumption when tracking a moving target, thus prolonging the network lifetime.

show abstract

“…The quantity

is a small value, controlling the convergence of the membership

. In our experiments,

, that is, our results obtained converged to the 0.01 specification level for comparing the two different methods [ 32 , 33 ]. The simulation lasted for 5000 time steps.…”

Section: Performance Evaluationmentioning

confidence: 62%

An Energy-Efficient Clustering Method for Target Tracking Based on Tracking Anchors in Wireless Sensor Networks

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Based on the prioritization, we use LOC-based cumulative-lift charts for evaluating the usefulness of the prioritization. The LOC-based cumulative lift chart is a commonly-used graph to evaluate the cost-effectiveness of defect prediction results [2] [8] [15]. In this chart, the x-axis considered as the requred test effort and the y-axis is the maximum number of discoverable defects by the assigned test effort [9] The x-axis denotes the cumulative lines of code (LOC) of selected modules, and the y-axis is the cumulative number of defects in the selected modules.…”

Section: Rule Prioritization and Its Evaluationmentioning

confidence: 99%

A Cost-Effectiveness Metric for Association Rule Mining in Software Defect Prediction

Nishiura

Kasagi

Monden

2022

Preprint

View full text Add to dashboard Cite

This paper proposes a cost-effectiveness metric for association rule mining suitable for software defect prediction where conditions of defective modules are characterized as association rules.Given a certain amount of test effort (or the number of test cases), the proposed metric is the expected number of defects to be discovered in modules that meet an association rule.Since the amount test effort is limited in general and full testing of all modules is ineffective, the proposed metric is useful to focus on the most cost effective set of modules to be tested with limited test effort. The proposed metric is defined based on the exponential Software Reliability Growth Model (SRGM) extended with the module size parameter, assuming that a larger module require more effort to discover defects. To evaluate the effectiveness of the proposed metric, association rules were extracted and prioritized based on the proposed metric using data sets of four open source software projects. The LOC-based cumulative-lift chart, which is often used to evaluate the cost effectiveness of defect prediction, shows that the proposed metric can focus on the rules that can discover more defects than the conventional association rule metrics, confidence and odds ratio.

show abstract

“…Past studies investigated how to use defect prediction models, aka classifiers, to predict the defectiveness of different types of entities including commits (Fan et al 2021;Giger et al 2012;Rodríguez-Pérez et al 2020;Tu et al 2020), classes Bangash et al 2020;Chen et al 2020;Chi et al 2017;Herbold et al , 2019Hosseini et al 2019;Jiarpakdee et al 2020;Lee et al 2016;Liu et al 2017;Nucci et al 2018;Qu et al 2021a;Shepperd et al 2018;Tantithamthavorn et al 2016cYan et al 2017; or methods ) by leveraging, for example, product metrics (Basili et al 1996;Gyimóthy et al 2005;Khoshgoftaar et al 1996;Nagappan and Ball 2005;Hassan 2009), process metrics (Moser et al 2008), knowledge from where previous defects occurred (Ostrand et al 2005;Kim et al 2007), information about change-inducing fixes (Kim et al 2008;Fukushima et al 2014) and, recently, deep learning techniques to automatically engineer features from source code elements .…”

Section: Introductionmentioning

confidence: 99%

On effort-aware metrics for defect prediction

2022

View full text Add to dashboard Cite

Context Advances in defect prediction models, aka classifiers, have been validated via accuracy metrics. Effort-aware metrics (EAMs) relate to benefits provided by a classifier in accurately ranking defective entities such as classes or methods. PofB is an EAM that relates to a user that follows a ranking of the probability that an entity is defective, provided by the classifier. Despite the importance of EAMs, there is no study investigating EAMs trends and validity. Aim The aim of this paper is twofold: 1) we reveal issues in EAMs usage, and 2) we propose and evaluate a normalization of PofBs (aka NPofBs), which is based on ranking defective entities by predicted defect density. Method We perform a systematic mapping study featuring 152 primary studies in major journals and an empirical study featuring 10 EAMs, 10 classifiers, two industrial, and 12 open-source projects. Results Our systematic mapping study reveals that most studies using EAMs use only a single EAM (e.g., PofB20) and that some studies mismatched EAMs names. The main result of our empirical study is that NPofBs are statistically and by orders of magnitude higher than PofBs. Conclusions In conclusion, the proposed normalization of PofBs: (i) increases the realism of results as it relates to a better use of classifiers, and (ii) promotes the practical adoption of prediction models in industry as it shows higher benefits. Finally, we provide a tool to compute EAMs to support researchers in avoiding past issues in using EAMs.

show abstract

Using K-core Decomposition on Class Dependency Networks to Improve Bug Prediction Model's Practical Performance

Cited by 47 publications

References 61 publications

An Energy-Efficient Clustering Method for Target Tracking Based on Tracking Anchors in Wireless Sensor Networks

An Energy-Efficient Clustering Method for Target Tracking Based on Tracking Anchors in Wireless Sensor Networks

A Cost-Effectiveness Metric for Association Rule Mining in Software Defect Prediction

On effort-aware metrics for defect prediction

Contact Info

Product

Resources

About