A5086461721- Advanced Decision-Making Techniques
- Traffic Prediction and Management Techniques
- Quantum Computing Algorithms and Architecture
- Topic Modeling
- Evaluation and Optimization Models
- Energy Efficient Wireless Sensor Networks
- Energy Harvesting in Wireless Networks
- Transportation Planning and Optimization
- Educational and Technological Research
- Advanced Technologies in Various Fields
- Neural Networks and Applications
- DNA and Biological Computing
- Model Reduction and Neural Networks
- Chaos-based Image/Signal Encryption
- Energy Load and Power Forecasting
- Molecular Communication and Nanonetworks
- Evaluation Methods in Various Fields
- Natural Language Processing Techniques
- Traffic control and management
- Quantum Information and Cryptography
- Time Series Analysis and Forecasting
- Metaheuristic Optimization Algorithms Research
- Biomedical Text Mining and Ontologies
- Fractional Differential Equations Solutions
Chongqing Normal University
2020-2022
Nanjing University of Posts and Telecommunications
2021-2022
Emotion recognition is a research hotspot in the field of artificial intelligence. If human-computer interaction system can sense human emotion and express emotion, it will make between robot more natural. In this paper, multimodal model based on many-objective optimization algorithm proposed for first time. The integrates voice information facial simultaneously optimize accuracy uniformity recognition. This paper compares with single-modal ISMS_ALA by recent related research. experimental...
Neural network has good self‐learning and adaptive capabilities. In this paper, a wavelet neural is proposed to be used solve the value problem of fractional differential equations (FDE). We construct (WNN) with structure 1 ×N× based on function give conditions for convergence given algorithm. This method uses truncated power series solution transform original equation into an approximate solution, then, using WNN, update parameters, finally get FDE solution. Simulation results prove validity WNN.
In order to solve the NP-hard problem of mobile sink path planning in wireless sensor networks (WSN) where communication range is modeled as a circular area and overlaps with each other, this paper proposes node method guided by Hamiltonian quantum annealing algorithm (EMGH) balance energy consumption networks, improve network life hole problem. First all, analyzes theory, transforms characteristics into mathematical model. The model considers that itself travelling salesman (TSP), but also...
Aiming at the problem of aerial hydrazine fuel guarantee safety assessment, an assessment model based on Vague-SPA and fuzzy close degree was proposed in accordance to matter-element.Firstly, definition actual situation, index system matter-element were built.Furthermore, weight optimized by applying decision model.The authors built with clustering combined using mathematics SPA analysis reduce influence subjectivity during evaluating process.Finally, objective reasonable results obtained...
On the basis of constructing safety assessment index system aerial ground support, natural weight and variable theory are introduced according with dynamic fuzzy nature index. The can be calculated adjusted to solve problem effectively reducing risk grade because some lower being neutralized by others. Then using extensible method make connection normalized, subordinate degree matrix positive negative fields level is set up based on both-branch model, in order optimize show importance factor...
The prediction of traffic flow plays a key role in the planning, control and management system. This paper proposed method including road correlation analysis based on Pearson coefficient (PCC) model which is composed convolutional neural network (CNN) gated recurrent unit (GRU) attention mechanism (PCC-AT-CNNGRU). Firstly, PCC applied to analyze spatial between roads select that related predicted construct spatial-temporal features matrix input it model. Then, CNN can mine feature flow, GRU...
Distributed representations are usually used as input features in text mining tasks. Previous works have shown its potential encoding semantics. Generally, there existing two representation methods namely static and dynamic, which means they context-free context-dependent respectively. Many demonstrated that context based significantly improved performance natural language processing field. Therefore, this paper, we utilize contextualized to recognize biomedical entities evaluate the results...