A5081837918- Service-Oriented Architecture and Web Services
- Advanced Software Engineering Methodologies
- Recommender Systems and Techniques
- Caching and Content Delivery
- Data Management and Algorithms
- Anomaly Detection Techniques and Applications
- Software System Performance and Reliability
- Web Data Mining and Analysis
- Domain Adaptation and Few-Shot Learning
- Multimodal Machine Learning Applications
- Advanced Database Systems and Queries
- Image Retrieval and Classification Techniques
- Semantic Web and Ontologies
- Cloud Computing and Resource Management
- Data Mining Algorithms and Applications
- Software Reliability and Analysis Research
- Human Pose and Action Recognition
- Machine Learning and Algorithms
- Advanced Image and Video Retrieval Techniques
- Data Quality and Management
- Big Data and Business Intelligence
- Scientific Computing and Data Management
- Video Surveillance and Tracking Methods
- Machine Learning and Data Classification
- Biomedical Text Mining and Ontologies
Hunan University
2025
Xiamen University
2025
Shanghai Institute of Optics and Fine Mechanics
2025
Rochester Institute of Technology
2015-2024
China University of Petroleum, East China
2023-2024
Beijing Jiaotong University
2018-2024
Zhejiang University
2022-2024
Jiangsu Normal University
2024
Changsha University
2024
Sichuan Normal University
2024
Mapping out the challenges and strategies for widespread adoption of service computing.
Web services are integrated software components for the support of interoperable machine-to-machine interaction over a network. have been widely employed building service-oriented applications in both industry and academia recent years. The number publicly available is steadily increasing on Internet. However, this proliferation makes it hard user to select proper service among large amount candidates. An inappropriate selection may cause many problems (e.g., ill-suited performance)...
The performance of a service provider may fluctuate due to the dynamic environment. Thus, quality actually delivered by is inherently uncertain. Existing optimization approaches usually assume that does not change over time. Moreover, most these rely on computing predefined objective function. When multiple criteria are considered, users required express their preference different (and sometimes conflicting) attributes as numeric weights. This rather demanding task and an imprecise...
Traffic accident anticipation aims to predict accidents from dashcam videos as early possible, which is critical safety-guaranteed self-driving systems. With cluttered traffic scenes and limited visual cues, it of great challenge how long there will be an observed frames. Most existing approaches are developed learn features accident-relevant agents for anticipation, while ignoring the their spatial temporal relations. Besides, current deterministic deep neural networks could overconfident...
In a real-world scenario, human actions are typically out of the distribution from training data, which requires model to both recognize known and reject unknown. Different image video more challenging be recognized in an open-set setting due uncertain temporal dynamics static bias actions. this paper, we propose Deep Evidential Action Recognition (DEAR) method open testing set. Specifically, formulate action recognition problem evidential deep learning (EDL) perspective novel calibration...
Service composition is emerging as an effective vehicle for integrating existing web services to create value-added and personalized composite services. As with similar functionality are expected be provided by competing providers, a key challenge find the "best" participate in composition. When multiple quality aspects (e.g., response time, fee, etc.) considered, weighting mechanism usually adopted most approaches, which requires users specify their preferences numeric values. We propose...
We present in this paper a novel QoS prediction approach to tackle service recommendation, which is recommend services with the best users. exploits available information estimate users' experience from previously unknown services. In regard, it can be modeled as general matrix completion problem, recover large small subset of entires. The infinite number ways complete an arbitrary makes problem extremely ill posed. highly sparse data further complicates challenges. Nonetheless, real-world...
The topological landscape of gene interaction networks provides a rich source information for inferring functional patterns genes or proteins. However, it is still challenging task to aggregate heterogeneous biological such as expression and interactions achieve more accurate inference prediction discovery new interactions. In particular, how generate unified vector representation integrate diverse input data key challenge addressed here. We propose scalable robust deep learning framework...
Background: As smart and automated applications pervade our lives, an increasing number of software developers are required to incorporate machine learning (ML) techniques into application development. However, acquiring the ML skill set can be nontrivial for owing both breadth depth domain. Aims: We seek understand challenges face in process development offer insights simplify process. Despite its importance, there has been little research on this topic. A few existing studies with...
Multiple-instance learning (MIL) provides an effective way to tackle the video anomaly detection problem by modeling it as a weakly supervised labels are usually only available at level while missing for frames due expensive labeling cost. We propose conduct novel Bayesian non-parametric submodular partition (BN-SVP) significantly improve MIL model training that can offer highly reliable solution robust in practical settings include outlier segments or multiple types of abnormal events....
We present a query algebra that supports optimized access of Web services through service-oriented queries. The service is defined based on formal model provides high-level abstraction across an application domain. defines set algebraic operators. Algebraic queries can be formulated using these This allows users to their desired both functionality and quality. provide the implementation each operator. enables generation Service Execution Plans (SEPs) used by directly services. optimization...
Classifying Web services and labeling them based on their functional features have played a major role in several fundamental service management tasks, such as discovery, selection, ranking, recommendation. Existing approaches leverage text mining techniques follow supervised learning process, which involves building classifier from training set of applying the to other services. This process requires intensive human effort set. In this paper, we propose idea pool-based active realize...
Abstract Background One of the most challenging tasks for bladder cancer diagnosis is to histologically differentiate two early stages, non-invasive Ta and superficially invasive T1, latter which associated with a significantly higher risk disease progression. Indeed, in considerable number cases, T1 tumors look very similar under microscope, making distinction difficult even experienced pathologists. Thus, there an urgent need favoring system based on machine learning (ML) distinguish...
Recent years have seen a surge in research on dynamic graph representation learning, which aims to model temporal graphs that are and evolving constantly over time. However, current work typically models dynamics with recurrent neural networks (RNNs), making them suffer seriously from computation memory overheads large graphs. So far, scalability of learning remains one the major challenges. In this paper, we present scalable framework, namely SpikeNet, efficiently capture structural...
As the number of Cloud services is growing at a tremendous speed, there an increasing service providers offering similar functionalities. Selecting with user desired non-functional properties (NFPs) becomes significant importance but triggers Big Data related research issues. First, selection decision should deal large volume NFPs data. Second, needs to reflect diverse preferences, including both qualitative and quantitative ones. Third, uncertainty network load leads high variability in...
Service-oriented architecture is a widely used software engineering paradigm to cope with complexity and dynamics in enterprise applications. Service composition, which provides cost-effective way implement systems, has attracted significant attention from both industry research communities. As online services may keep evolving over time thus lead highly dynamic environment, service composition must be self-adaptive tackle uninformed behavior during the evolution of services. In addition,...