Joint feature representation learning and progressive distribution matching for cross-project defect prediction

2021 ◽  
Vol 137 ◽  
pp. 106588
Author(s):  
Quanyi Zou ◽  
Lu Lu ◽  
Zhanyu Yang ◽  
Xiaowei Gu ◽  
Shaojian Qiu
Keyword(s):  
Representation Learning ◽  
Feature Representation ◽  
Defect Prediction ◽  
Distribution Matching ◽  
Cross Project

scholarly journals Transfer Convolutional Neural Network for Cross-Project Defect Prediction

2019 ◽  
Vol 9 (13) ◽  
pp. 2660 ◽  
Author(s):  
Shaojian Qiu ◽  
Hao Xu ◽  
Jiehan Deng ◽  
Siyu Jiang ◽  
Lu Lu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Reproducing Kernel ◽  
Reproducing Kernel Hilbert Space ◽  
Defect Prediction ◽  
Classification Error ◽  
Software Defect ◽  
Novel Approach ◽  
Distribution Matching ◽  
Cross Project

Cross-project defect prediction (CPDP) is a practical solution that allows software defect prediction (SDP) to be used earlier in the software lifecycle. With the CPDP technique, the software defect predictor trained by labeled data of mature projects can be applied for the prediction task of a new project. Most previous CPDP approaches ignored the semantic information in the source code, and existing semantic-feature-based SDP methods do not take into account the data distribution divergence between projects. These limitations may weaken defect prediction performance. To solve these problems, we propose a novel approach, the transfer convolutional neural network (TCNN), to mine the transferable semantic (deep-learning (DL)-generated) features for CPDP tasks. Specifically, our approach first parses the source file into integer vectors as the network inputs. Next, to obtain the TCNN model, a matching layer is added into convolutional neural network where the hidden representations of the source and target project-specific data are embedded into a reproducing kernel Hilbert space for distribution matching. By simultaneously minimizing classification error and distribution divergence between projects, the constructed TCNN could extract the transferable DL-generated features. Finally, without losing the information contained in handcrafted features, we combine them with transferable DL-generated features to form the joint features for CPDP performing. Experiments based on 10 benchmark projects (with 90 pairs of CPDP tasks) showed that the proposed TCNN method is superior to the reference methods.

Heterogeneous Cross Project Defect Prediction in Software

2020 ◽  
Author(s):  
Sonali Srivastava ◽  
Shikha Rani ◽  
Shailly Singh ◽  
Saurabh Singh ◽  
Rohit Vashisht
Keyword(s):  
Defect Prediction ◽  
Cross Project

An Improved SDA Based Defect Prediction Framework for Both Within-Project and Cross-Project Class-Imbalance Problems

2017 ◽  
Vol 43 (4) ◽  
pp. 321-339 ◽  
Author(s):  
Xiao-Yuan Jing ◽  
Fei Wu ◽  
Xiwei Dong ◽  
Baowen Xu
Keyword(s):  
Class Imbalance ◽  
Defect Prediction ◽  
Cross Project

Cross-project smell-based defect prediction

2021 ◽  
Author(s):  
Bruno Sotto-Mayor ◽  
Meir Kalech
Keyword(s):  
Defect Prediction ◽  
Cross Project

scholarly journals HDA: Cross-Project Defect Prediction via Heterogeneous Domain Adaptation With Dictionary Learning

IEEE Access ◽  
2018 ◽  
Vol 6 ◽  
pp. 57597-57613 ◽  
Author(s):  
Zhou Xu ◽  
Peipei Yuan ◽  
Tao Zhang ◽  
Yutian Tang ◽  
Shuai Li ◽  
...  
Keyword(s):  
Dictionary Learning ◽  
Domain Adaptation ◽  
Defect Prediction ◽  
Cross Project

An empirical study of just-in-time defect prediction using cross-project models

2014 ◽  
Author(s):  
Takafumi Fukushima ◽  
Yasutaka Kamei ◽  
Shane McIntosh ◽  
Kazuhiro Yamashita ◽  
Naoyasu Ubayashi
Keyword(s):  
Empirical Study ◽  
Defect Prediction ◽  
Just In Time ◽  
Cross Project
Sign in / Sign up

Export Citation Format

Share Document