Software defect prediction plays a critical role in software development and quality assurance processes. Effective defect prediction enables testers to accurately prioritize testing efforts and enhance defect detecti...Software defect prediction plays a critical role in software development and quality assurance processes. Effective defect prediction enables testers to accurately prioritize testing efforts and enhance defect detection efficiency. Additionally, this technology provides developers with a means to quickly identify errors, thereby improving software robustness and overall quality. However, current research in software defect prediction often faces challenges, such as relying on a single data source or failing to adequately account for the characteristics of multiple coexisting data sources. This approach may overlook the differences and potential value of various data sources, affecting the accuracy and generalization performance of prediction results. To address this issue, this study proposes a multivariate heterogeneous hybrid deep learning algorithm for defect prediction (DP-MHHDL). Initially, Abstract Syntax Tree (AST), Code Dependency Network (CDN), and code static quality metrics are extracted from source code files and used as inputs to ensure data diversity. Subsequently, for the three types of heterogeneous data, the study employs a graph convolutional network optimization model based on adjacency and spatial topologies, a Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) hybrid neural network model, and a TabNet model to extract data features. These features are then concatenated and processed through a fully connected neural network for defect prediction. Finally, the proposed framework is evaluated using ten promise defect repository projects, and performance is assessed with three metrics: F1, Area under the curve (AUC), and Matthews correlation coefficient (MCC). The experimental results demonstrate that the proposed algorithm outperforms existing methods, offering a novel solution for software defect prediction.展开更多
Spatial information remains to be an important topic in geographic information system and in remote sensing fields,and spatial relationships have been increasingly incorporated into the image classification processes....Spatial information remains to be an important topic in geographic information system and in remote sensing fields,and spatial relationships have been increasingly incorporated into the image classification processes.Previous studies have employed multiple occurrences of spatial features(shape,texture,etc.,)to improve classification results.However,less attention has been focused on using higher-level spatial relationships for image classification.In this study,two novel spatial relationships,namely,maximum spatial adjacency(MSA)and directional spatial adjacency(DSA),were proposed to assist in image classification.The proposed methods were implemented to extract buildings,beach,and emergent vegetation land-cover classes according to their spatial relationships with their corresponding reference classes.The promising results obtained from this study suggest that the proposed MSA and DSA spatial relationships can be valuable information in defining rule sets for a more reasonable and accurate classification.展开更多
文摘Software defect prediction plays a critical role in software development and quality assurance processes. Effective defect prediction enables testers to accurately prioritize testing efforts and enhance defect detection efficiency. Additionally, this technology provides developers with a means to quickly identify errors, thereby improving software robustness and overall quality. However, current research in software defect prediction often faces challenges, such as relying on a single data source or failing to adequately account for the characteristics of multiple coexisting data sources. This approach may overlook the differences and potential value of various data sources, affecting the accuracy and generalization performance of prediction results. To address this issue, this study proposes a multivariate heterogeneous hybrid deep learning algorithm for defect prediction (DP-MHHDL). Initially, Abstract Syntax Tree (AST), Code Dependency Network (CDN), and code static quality metrics are extracted from source code files and used as inputs to ensure data diversity. Subsequently, for the three types of heterogeneous data, the study employs a graph convolutional network optimization model based on adjacency and spatial topologies, a Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) hybrid neural network model, and a TabNet model to extract data features. These features are then concatenated and processed through a fully connected neural network for defect prediction. Finally, the proposed framework is evaluated using ten promise defect repository projects, and performance is assessed with three metrics: F1, Area under the curve (AUC), and Matthews correlation coefficient (MCC). The experimental results demonstrate that the proposed algorithm outperforms existing methods, offering a novel solution for software defect prediction.
基金This research is partially supported by a NSERC Discovery Grant awarded to Dr.Jinfei Wang,University of Western Ontario.
文摘Spatial information remains to be an important topic in geographic information system and in remote sensing fields,and spatial relationships have been increasingly incorporated into the image classification processes.Previous studies have employed multiple occurrences of spatial features(shape,texture,etc.,)to improve classification results.However,less attention has been focused on using higher-level spatial relationships for image classification.In this study,two novel spatial relationships,namely,maximum spatial adjacency(MSA)and directional spatial adjacency(DSA),were proposed to assist in image classification.The proposed methods were implemented to extract buildings,beach,and emergent vegetation land-cover classes according to their spatial relationships with their corresponding reference classes.The promising results obtained from this study suggest that the proposed MSA and DSA spatial relationships can be valuable information in defining rule sets for a more reasonable and accurate classification.
