A5036273604- Data Management and Algorithms
- Topic Modeling
- Advanced Database Systems and Queries
- Hydrological Forecasting Using AI
- Cloud Computing and Resource Management
- Water Quality Monitoring Technologies
- Natural Language Processing Techniques
- Data Quality and Management
- Hydrology and Watershed Management Studies
- Advanced Text Analysis Techniques
- Graph Theory and Algorithms
- Flood Risk Assessment and Management
- Geographic Information Systems Studies
- Distributed and Parallel Computing Systems
- Flow Measurement and Analysis
- Caching and Content Delivery
- Semantic Web and Ontologies
- Automated Road and Building Extraction
- Advanced Graph Neural Networks
- Web Data Mining and Analysis
- Parallel Computing and Optimization Techniques
- Network Traffic and Congestion Control
- Model Reduction and Neural Networks
- Advanced Chemical Sensor Technologies
- Image and Video Quality Assessment
Guilin University of Technology
2024-2025
Shandong University of Traditional Chinese Medicine
2025
Hohai University
2015-2024
Ministry of Water Resources of the People's Republic of China
2021-2024
China Agricultural University
2022-2024
North University of China
2024
Ministry of Agriculture and Rural Affairs
2022-2024
National Engineering Research Center for Information Technology in Agriculture
2022-2024
Lanzhou University of Technology
2024
Southwest University of Science and Technology
2024
Machine learning models, particularly reinforcement (RL), have demonstrated great potential in optimizing video streaming applications. However, the state-of-the-art solutions are limited to an "offline learning" paradigm, i.e., RL models trained simulators and then operated real networks. As a result, they inevitably suffer from simulation-to-reality gap, showing far less satisfactory performance under conditions compared with simulated environment. In this work, we close gap by proposing...
Hadoop is an efficient and simple parallel framework following the Map Reduce paradigm, making processing recently become a hot issue in data-intensive applications. Since can be easily deployed on large-scale clusters including up to thousands of computers, various studies intend process common relational database operations also this new platform expect achieve remarkable performance. However, these works have prepare customized programs according different input format, communication...
Efficient scheduling algorithms are key for attaining high performance in heterogeneous computing systems. In this article, we propose a new list algorithm assigning task graphs to fully connected processors with an aim minimize the length. The proposed algorithm, called Improved Predict Priority Task Scheduling (IPPTS) has two phases: prioritization phase, which gives priority tasks, and processor selection selects task. IPPTS quadratic time complexity as related same goal, that is...
During the process of smart city construction, managers always spend a lot energy and money for cleaning street garbage due to random appearances garbage. Consequently, visual cleanliness assessment is particularly important. However, existing approaches have some clear disadvantages, such as collection information not automated real-time. To address these this paper proposes novel urban approach using mobile edge computing deep learning. First, high-resolution cameras installed on vehicles...
This study aims to predict and diagnose pediatric septic shock through the screening of immune infiltration-related biomarkers. Three gene expression datasets were accessible from Gene Expression Omnibus repository. The differentially expressed genes identified using R 4.3.2 ( https://www.r-project.org/ ), followed by set enrichment analysis. Thereafter, utilizing machine-learning algorithms. receiver operating characteristic curve was employed assess discrimination effectiveness hub genes....
Fish species identification plays a vital role in marine fisheries resource exploration, yet datasets related to fish resources are scarce. In open-water environments, various often exhibit similar appearances and sizes. To solve these issues, we propose few-shot learning approach identifying species. Our involves two key components. Firstly, the embedding module was designed address challenges posed by large number of with phenotypes utilizing distribution relationships space. Secondly,...
This paper presents a list-based scheduling algorithm called Predict Priority Task Scheduling (PPTS) for heterogeneous computing. The main goal is to minimize the length by introducing lookahead feature in two phases of PPTS algorithm, namely task prioritizing phase and processor selection phase. Existing list algorithms, such as PEFT Lookahead have introduced this only novelty its ability look ahead not but also phase, without increasing time complexity. achieved based on predict cost...
Parallel Secondo scales up the capability of processing extensible data models in Secondo. It combines Hadoop with a set databases, providing almost all existing SECONDO types and operators. Therefore it is possible for user to convert large-scale sequential queries parallel without learning Map/Reduce programming details. This paper demonstrates such procedure. imports from project OpenStreetMap into databases build urban traffic network then processes network-based like map-matching...
This paper presents a hybrid parallel processing system, named Parallel Secondo. It combines the Hadoop framework and set of single-computer Secondo databases, in order to introduce mobility data procedures into community, vice versa. The system keeps front-end executable language allow users state their queries like common sequential queries. Besides, auxiliary scripts is provided so as make it easier manage no matter how large underlying cluster is, keep platform transparent level system....
With the rapid advancement of deep learning techniques, learning-based flood prediction models have drawn significant attention. However, for short-term in small- and medium-sized river basins, typically rely on hydrological data spanning from past several hours to one day, utilizing a fixed-length input window. Such limits models’ adaptability diverse rainfall events restricts their capability comprehend historical temporal patterns. To address underutilization information by existing...
Secondo is an extensible DBMS prototype. It emphasizes the handling of spatial and moving objects data provides sophisticated models query processing over such data, offering specialized types operations. More generally, there a wealth techniques available, for example, (i) construction road network from OpenStreetMap map matching trajectories against this network, (ii) symbolic with pattern indexing techniques, (iii) spatio-temporal queries, (iv) similarity search on M-trees, (v) combining...
With the recent advancements in information and communication technologies, large number of devices are connecting to Internet, hence volumes data different formats from sources generating. Consequently, on one hand dynamic heterogeneous sharing management, ecosystem Internet Things (IoT), where every smart object is connected presents new research challenges. On other hand, citizen privacy preserving another challenge, because he/she has send his/her a service provider, obtain required...
Recent years have witnessed a rise of learning-based ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , artificial intelligence driven or AI-driven) video transport design, in order to achieve consistently high performance, even when the modern Internet is becoming increasingly heterogeneous while applications are unprecedentedly demanding simultaneous high-throughput and low-latency requirements HD telephony intelligent remote...
Data-driven models have been successfully applied to flood prediction. However, the nonlinearity and uncertainty of prediction process possible noise or outliers in data set will lead incorrect results. In addition, data-driven are only trained from available datasets do not involve scientific principles laws during model training process, which may predictions that conform physical laws. To this end, we propose a method based on knowledge-guided heterogeneous graphs temporal convolutional...