A5102770466- Software Engineering Research
- Advanced Clustering Algorithms Research
- Imbalanced Data Classification Techniques
- Data Management and Algorithms
- Text and Document Classification Technologies
- Software Testing and Debugging Techniques
- Geoscience and Mining Technology
- Blind Source Separation Techniques
- Multimedia Communication and Technology
- Network Security and Intrusion Detection
- Advanced Malware Detection Techniques
- Software System Performance and Reliability
- Respiratory Support and Mechanisms
- Hyperglycemia and glycemic control in critically ill and hospitalized patients
- Evaluation Methods in Various Fields
- Military Defense Systems Analysis
- Image and Signal Denoising Methods
- Electrolyte and hormonal disorders
- Complex Network Analysis Techniques
- Machine Learning and Data Classification
- Computational Physics and Python Applications
- Advanced Algorithms and Applications
- Software-Defined Networks and 5G
- Video Analysis and Summarization
- Algorithms and Data Compression
Beijing University of Posts and Telecommunications
2024
Switch
2024
Shandong University
2006-2023
Dalian Maritime University
2018-2020
Tianjin University
2007-2008
Due to the unavoidable bugs appearing in most of software systems, bug resolution has become one important activities maintenance. To decrease time cost manual work, text classification techniques are applied automatically identify severity reports. In this paper, we address problem low-quality and class imbalance for identifying First, combine feature selection with instance simultaneously reduce report dimension word dimension, which could get small-scale high-quality reduced data set....
Current cooperative multi-agent algorithms consider the sum of team rewards as joint objective. These assumes a consistent monotonicity between individual value functions and functions. However, constraints are unable to accurately fit scenarios in autonomous search where (1) Agents influence each other during action execution (2) Individual conflict with rewards. Hence, we proposed method hierarchical policy constrains. Locally, avoid path conflicts caused by agent policies, use weighted...
In this paper, we drive the wavelet threshold error function, adopt particle swarm optimization (PSO) and genetic algorithm (GA) to determine optimal denoising threshold, results show that our methods have better performance in signal processing.
As software systems become increasingly large, the logic becomes more complex, resulting in a large number of bug reports being submitted to repository daily. Due tight schedules and limited human resources, developers may not have enough time inspect all bugs. Thus, they often concentrate on bugs that impacts. However, there are two main challenges limiting automation technology would help aware high-impact early, namely, low quality class distribution imbalance. To address these...
Since a large number of bug reports are submitted to the repository every day, efficiently assigning correct developer is considerable challenge. Because differences between different components projects, current classification mainly relies on report dispatch designated or community. Unfortunately, component information filled in by default according submitter and result often incorrect. Thus, an automatic technology that can identify high-impact help developers be aware them early, rectify...
Acute respiratory distress syndrome (ARDS) is a serious and progressive clinical complication for which there currently no effective pharmacological treatment. Inflammatory responses pulmonary surfactant dysregulation are key features of ARDS pathogenesis. Levosimendan, calcium sensitizer used as an inodilator to treat decompensated heart failure, may have several beneficial properties in ARDS, including vasodilation, anti-inflammatory, cardioprotective properties. However, the therapeutic...
Despite the great potential to save labor cost of developers, automated bug triaging as a text classification problem has not been thoroughly investigated on long descriptions, which are informative but often noisy.In this paper an effective triage technique is proposed build high quality set data by removing noisy and noninformative reports while assigning new bugs appropriate developer.The techniqueweighted reductionis built upon three feature selection algorithms four instances with...