Cross Project Defect Prediction via Balanced Distribution Adaptation Based Transfer Learning

Zhou, Xu; Pang, Shaoning; Zhang, Tao; Luo, Xiapu; Liu, Jin; Tang, Yutian; Yu, Xiao; Xue, Lei

doi:10.1007/s11390-019-1959-z

Cited by 59 publications

(51 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…0.05/8 = 0.00625) to compute the significant difference of the model performance. This evaluation has been widely used for performance comparison in many defect prediction studies [38, 45, 58–60]. Friedman test is used to determine if there are statistically significant differences between multiple models, and Nemenyi test is performed to check which model differs significantly.…”

Section: Experiments Resultsmentioning

confidence: 99%

Manifold embedded distribution adaptation for cross‐project defect prediction

Sun

Jing

et al. 2020

IET softw.

View full text Add to dashboard Cite

Cross‐project defect prediction (CPDP) technology refers to the constructing prediction model to predict the instance label of the target project by utilising labelled data from an external project. The challenge of CPDP methods is the distribution difference between the data from different projects. Transfer learning can transfer the knowledge from the source domain to the target domain with the aim to minimise the domain difference between different domains. However, most existing methods reduce the distribution discrepancy in the original feature space, where the features are high‐dimensional and non‐linear, which makes it hard to reduce the distribution distance between different projects. Moreover, previous works mainly consider marginal distribution or conditional distribution difference. In this study, the authors proposed a manifold embedded distribution adaptation (MDA) approach to narrow the distribution gap in manifold feature subspace. MDA maps source and target project data to manifold subspace and then joint distribution adaptation of conditional and marginal distributions is performed on manifold subspace. To evaluate the effectiveness of MDA, the authors perform extensive experiments on 20 public projects with three indicators. The experiment results show that MDA improves the average performance, but the improvement is not statistically significant in comparison to HYDRA (one of the baselines).

show abstract

Section: Experiments Resultsmentioning

confidence: 99%

Manifold embedded distribution adaptation for cross‐project defect prediction

Sun

Jing

et al. 2020

IET softw.

View full text Add to dashboard Cite

show abstract

“…To simultaneously consider both the marginal distribution and conditional distribution, Qiu et al [44] proposed joint distribution matching (JDM) to construct new feature representation that is effective and robust for substantial distribution difference for CPDP. Different from JDM, balanced distribution adaption (BDA) methods proposed by Xu et al [45] not only take the marginal distribution and conditional distribution into account but also adaptively assign different weights to them.…”

Section: Related Work a Software Defect Predictionmentioning

confidence: 99%

Semantic Feature Learning via Dual Sequences for Defect Prediction

Lin

2021

IEEE Access

View full text Add to dashboard Cite

Software defect prediction (SDP) can help developers reasonably allocate limited resources for locating bugs and prioritizing their testing efforts. Existing methods often serialize an Abstract Syntax Tree (AST) obtained from the program source code into a token sequence, which is then inputted into the deep learning model to learn the semantic features. However, there are different ASTs with the same token sequence, and it is impossible to distinguish the tree structure of the ASTs only by a token sequence. To solve this problem, this paper proposes a framework called Semantic Feature Learning via Dual Sequences (SFLDS), which can capture the semantic and structural information in the AST for feature generation. Specifically, based on the AST, we select the representative nodes in the AST and convert the program source code into a simplified AST (S-AST). Our method introduces two sequences to represent the semantic and structural information of the S-AST, one is the result of traversing the S-AST node in pre-order, and another is composed of parent nodes. Then each token in the dual sequences is encoded as a numerical vector via mapping and word embedding. Finally, we use a bi-directional long short-term memory (BiLSTM) based neural network to automatically generate semantic features from the dual sequences for SDP. In addition, to leverage the statistical characteristics contained in the handcrafted metrics, we also propose a framework called Defect Prediction via SFLDS (DP-SFLDS) which combines the semantic features generated from SFLDS with handcrafted metrics to perform SDP. In our empirical studies, eight open-source Java projects from the PROMISE repository are chosen as our empirical subjects. Experimental results show that our proposed approach can perform better than several state-of-the-art baseline SDP methods.INDEX TERMS Software defect prediction, abstract syntax tree, deep learning, bi-directional long short-term memory network.LU LU received the Ph.D. degree from Xi'an Jiaotong University, in 1999. He is currently a Professor with the School of Computer Science and Engineering, South China University of Technology, China. His main research interests include software engineering, software testing, and software architecture design.

show abstract

“…Each project has its own features such as project size and number of bugs [3]. Identifying similarities among the datasets is the basis to eliminate differences between the data across projects.…”

Section: Similarity Scoringmentioning

confidence: 99%

“…Reliability is one of the most significant attributes to enhance the quality of the product in the software development process [1][2][3]. Assessing software reliability is vital to deliver failure free software system.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Cross-Project Software Reliability Growth Model using Project Similarity Based Clustering

San¹,

Washizaki²,

Fukazawa³

et al. 2021

Preprint

View full text Add to dashboard Cite

Software reliability is an important characteristic for ensuring the qualities of software products. Predicting the potential number of bugs from the beginning of a development project allows practitioners to make the appropriate decisions regarding testing activities. In the initial development phases, applying traditional software reliability growth models (SRGMs) with limited past data does not always provide reliable prediction result for decision making. To overcome this, herein we propose a new software reliability modeling method called deep cross-project software reliability growth model (DC-SRGM). DC-SRGM is a cross-project prediction method that uses features of previous projects’ data through project similarity. Specifically, the proposed method applies cluster-based project selection for training data source and modeling by a deep learning method. Experiments involving 15 real datasets from a company and 11 open source software datasets show that DC-SRGM can more precisely describe the reliability of ongoing development projects than existing traditional SRGMs and the LSTM model.

show abstract

Cross Project Defect Prediction via Balanced Distribution Adaptation Based Transfer Learning

Cited by 59 publications

References 53 publications

Manifold embedded distribution adaptation for cross‐project defect prediction

Manifold embedded distribution adaptation for cross‐project defect prediction

Semantic Feature Learning via Dual Sequences for Defect Prediction

Deep Cross-Project Software Reliability Growth Model using Project Similarity Based Clustering

Contact Info

Product

Resources

About