An Improved CNN Model for Within-Project Software Defect Prediction

Pan, Cong; Lu, Minyan; Xu, Biao; Gao, Houleng

doi:10.3390/app9102138

Cited by 95 publications

(74 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To further illustrate the effectiveness of the proposed algorithm and evaluate the network performance, the paper analyzed the time complexity [45,46] of the dual channels and compared them in turn with those of the improved network. The specific parameters are shown in Table 2.…”

Section: Efficiency Comparisonmentioning

confidence: 99%

RETRACTED: Image Super-Resolution Algorithm Based on Dual-Channel Convolutional Neural Networks

et al. 2019

View full text Add to dashboard Cite

For the image super-resolution method from a single channel, it is difficult to achieve both fast convergence and high-quality texture restoration. By mitigating the weaknesses of existing methods, the present paper proposes an image super-resolution algorithm based on dual-channel convolutional neural networks (DCCNN). The novel structure of the network model was divided into a deep channel and a shallow channel. The deep channel was used to extract the detailed texture information from the original image, while the shallow channel was mainly used to recover the overall outline of the original image. Firstly, the residual block was adjusted in the feature extraction stage, and the nonlinear mapping ability of the network was enhanced. The feature mapping dimension was reduced, and the effective features of the image were obtained. In the up-sampling stage, the parameters of the deconvolutional kernel were adjusted, and high-frequency signal loss was decreased. The high-resolution feature space could be rebuilt recursively using long-term and short-term memory blocks during the reconstruction stage, further enhancing the recovery of texture information. Secondly, the convolutional kernel was adjusted in the shallow channel to reduce the parameters, ensuring that the overall outline of the image was restored and that the network converged rapidly. Finally, the dual-channel loss function was jointly optimized to enhance the feature-fitting ability in order to obtain the final high-resolution image output. Using the improved algorithm, the network converged more rapidly, the image edge and texture reconstruction effect were obviously improved, and the Peak Signal-to-Noise Ratio (PSNR) and structural similarity were also superior to those of other solutions.

show abstract

Section: Efficiency Comparisonmentioning

confidence: 99%

RETRACTED: Image Super-Resolution Algorithm Based on Dual-Channel Convolutional Neural Networks

et al. 2019

View full text Add to dashboard Cite

show abstract

“…But, as mentioned before, public datasets can have inheriting issues, especially when it comes to quality, which can lead to poor prediction results. To solve this problem, we can consider applying some proper data cleaning and data pre-processing techniques (Pan et al, 2019).…”

Section: Discussionmentioning

confidence: 99%

“…In the future, we can improve the selection of bug prediction techniques using machine learning because merely using better technique than before does not guarantee the improvement of performance. Still, there are some researchers focused on proposing hybrid frameworks (Erturk and Sezer, 2015;Arar and Ayan, 2015;Rhmann et al, 2020;Miholca et al, 2018;Abaei et al, 2015;Ryu et al, 2015), or improving the existing techniques (Pan et al, 2019;Zhou et al, 2019;Rathore and Kumar, 2017;Wei et al, 2018).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Techniques for Software Bug Prediction: A Systematic Review

Saharudin¹,

Koh²,

Kew³

2020

Journal of Computer Science

View full text Add to dashboard Cite

The goal of software bug prediction is to identify the software modules that will have the likelihood to get bugs by using some fundamental project resources before the real testing starts. Due to high cost in correcting the detected bugs, it is advisable to start predicting bugs at the early stage of development instead of at the testing phase. There are many techniques and approaches that can be used to build the prediction models, such as machine learning. This technique is widely used nowadays because it can give accurate results and analysis. Therefore, we decided to perform a review of past literature on software bug prediction and machine learning so that we can understand better about the process of constructing the prediction model. Not only we want to see the machine learning techniques that past researchers used, we also assess the datasets, metrics and performance measures that are used during the development of the models. In this study, we have narrowed down to 31 main studies and six types of machine learning techniques have been identified. Two public datasets are found to be frequently used and object-oriented metrics are the highly chosen metrics for the prediction model. As for the performance measure, both graphical and numerical measures are often used to evaluate the performance of the models. From the results, we conclude that the machine learning technique can predict the bug, but there are not many applications in this area that exist nowadays. There are a few challenges in constructing the prediction model. Thus, more studies need to be carried out so that a well-formed result is obtained. We also provide a recommendation for future research based on the results we got from this study.

show abstract

“…Pan etc. [ 41 ] used the upgraded version of Convolutional Neural Network (CNN) by using the in-project SDP and evaluated 12 different versions of the project. They compared it with the SDP method based on baseline deep learning.…”

Section: Background and Related Workmentioning

confidence: 99%

Attention based GRU-LSTM for software defect prediction

et al. 2021

View full text Add to dashboard Cite

Software defect prediction (SDP) can be used to produce reliable, high-quality software. The current SDP is practiced on program granular components (such as file level, class level, or function level), which cannot accurately predict failures. To solve this problem, we propose a new framework called DP-AGL, which uses attention-based GRU-LSTM for statement-level defect prediction. By using clang to build an abstract syntax tree (AST), we define a set of 32 statement-level metrics. We label each statement, then make a three-dimensional vector and apply it as an automatic learning model, and then use a gated recurrent unit (GRU) with a long short-term memory (LSTM). In addition, the Attention mechanism is used to generate important features and improve accuracy. To verify our experiments, we selected 119,989 C/C++ programs in Code4Bench. The benchmark tests cover various programs and variant sets written by thousands of programmers. As an evaluation standard, compared with the state evaluation method, the recall, precision, accuracy and F1 measurement of our well-trained DP-AGL under normal conditions have increased by 1%, 4%, 5%, and 2% respectively.

show abstract

An Improved CNN Model for Within-Project Software Defect Prediction

Cited by 95 publications

References 53 publications

RETRACTED: Image Super-Resolution Algorithm Based on Dual-Channel Convolutional Neural Networks

RETRACTED: Image Super-Resolution Algorithm Based on Dual-Channel Convolutional Neural Networks

Machine Learning Techniques for Software Bug Prediction: A Systematic Review

Attention based GRU-LSTM for software defect prediction

Contact Info

Product

Resources

About