Siamese Dense Neural Network for Software Defect Prediction With Small Data
Software defect prediction (SDP) exerts a major role in software development, concerning reducing software costs and ensuring software quality. However, developing an accurate SDP model is still a severe and challenging task with the lack of training data. Fortunately, Siamese networks are powerful...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8585009/ |
id |
doaj-a585d568a8b24a44850546c09beb80a2 |
---|---|
record_format |
Article |
spelling |
doaj-a585d568a8b24a44850546c09beb80a22021-03-29T22:51:10ZengIEEEIEEE Access2169-35362019-01-0177663767710.1109/ACCESS.2018.28890618585009Siamese Dense Neural Network for Software Defect Prediction With Small DataLinchang Zhao0https://orcid.org/0000-0002-6502-587XZhaowei Shang1Ling Zhao2Anyong Qin3Yuan Yan Tang4College of Computer Science, Chongqing University, Chongqing, ChinaCollege of Computer Science, Chongqing University, Chongqing, ChinaUnited Imaging (Guizhou) Healthcare Co., Ltd, Guiyang, ChinaCollege of Computer Science, Chongqing University, Chongqing, ChinaFaculty of Science and Technology, University of Macau, Macau, ChinaSoftware defect prediction (SDP) exerts a major role in software development, concerning reducing software costs and ensuring software quality. However, developing an accurate SDP model is still a severe and challenging task with the lack of training data. Fortunately, Siamese networks are powerful for learning a few samples and have been perfectly used in other fields. This paper explores the advantages of Siamese networks to propose a novel SDP model, Siamese dense neural networks (SDNNs), which integrates similarity feature learning and distance metric learning into a unified approach. It mainly includes two phases: model building and training. To be more specific, it means building the novel SDNN for capturing the highest-level similarity features and training the model to realize prediction through the designed contrast loss function with cosine proximity. Importantly, we extensively compared the SDNN approach with the state-of-the-art SDP approaches utilizing 10 software defect datasets. The experimental results show that our SDNN is a competitive approach and is able to improve the prediction performance more significantly compared with the benchmarked approaches.https://ieeexplore.ieee.org/document/8585009/Siamese dense neural networksdeep learningmetric learningfew-shot learningsoftware defect prediction |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Linchang Zhao Zhaowei Shang Ling Zhao Anyong Qin Yuan Yan Tang |
spellingShingle |
Linchang Zhao Zhaowei Shang Ling Zhao Anyong Qin Yuan Yan Tang Siamese Dense Neural Network for Software Defect Prediction With Small Data IEEE Access Siamese dense neural networks deep learning metric learning few-shot learning software defect prediction |
author_facet |
Linchang Zhao Zhaowei Shang Ling Zhao Anyong Qin Yuan Yan Tang |
author_sort |
Linchang Zhao |
title |
Siamese Dense Neural Network for Software Defect Prediction With Small Data |
title_short |
Siamese Dense Neural Network for Software Defect Prediction With Small Data |
title_full |
Siamese Dense Neural Network for Software Defect Prediction With Small Data |
title_fullStr |
Siamese Dense Neural Network for Software Defect Prediction With Small Data |
title_full_unstemmed |
Siamese Dense Neural Network for Software Defect Prediction With Small Data |
title_sort |
siamese dense neural network for software defect prediction with small data |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
Software defect prediction (SDP) exerts a major role in software development, concerning reducing software costs and ensuring software quality. However, developing an accurate SDP model is still a severe and challenging task with the lack of training data. Fortunately, Siamese networks are powerful for learning a few samples and have been perfectly used in other fields. This paper explores the advantages of Siamese networks to propose a novel SDP model, Siamese dense neural networks (SDNNs), which integrates similarity feature learning and distance metric learning into a unified approach. It mainly includes two phases: model building and training. To be more specific, it means building the novel SDNN for capturing the highest-level similarity features and training the model to realize prediction through the designed contrast loss function with cosine proximity. Importantly, we extensively compared the SDNN approach with the state-of-the-art SDP approaches utilizing 10 software defect datasets. The experimental results show that our SDNN is a competitive approach and is able to improve the prediction performance more significantly compared with the benchmarked approaches. |
topic |
Siamese dense neural networks deep learning metric learning few-shot learning software defect prediction |
url |
https://ieeexplore.ieee.org/document/8585009/ |
work_keys_str_mv |
AT linchangzhao siamesedenseneuralnetworkforsoftwaredefectpredictionwithsmalldata AT zhaoweishang siamesedenseneuralnetworkforsoftwaredefectpredictionwithsmalldata AT lingzhao siamesedenseneuralnetworkforsoftwaredefectpredictionwithsmalldata AT anyongqin siamesedenseneuralnetworkforsoftwaredefectpredictionwithsmalldata AT yuanyantang siamesedenseneuralnetworkforsoftwaredefectpredictionwithsmalldata |
_version_ |
1724190695696105472 |