A5100343458- Air Quality Monitoring and Forecasting
- Landslides and related hazards
- Atmospheric chemistry and aerosols
- Spam and Phishing Detection
- Energy Load and Power Forecasting
- Hydrology and Watershed Management Studies
- Hydrological Forecasting Using AI
- Air Quality and Health Impacts
- Geotechnical and Geomechanical Engineering
- Misinformation and Its Impacts
- Image Processing Techniques and Applications
- Graph theory and applications
- Traffic Prediction and Management Techniques
- Simulation and Modeling Applications
- Additive Manufacturing Materials and Processes
- Reinforcement Learning in Robotics
- Analytical Chemistry and Sensors
- Generative Adversarial Networks and Image Synthesis
- Military Defense Systems Analysis
- Air Traffic Management and Optimization
- Finite Group Theory Research
- Adaptive Dynamic Programming Control
- graph theory and CDMA systems
- Engineering and Test Systems
- Urban and Freight Transport Logistics
Northwestern Polytechnical University
2012-2024
Chongqing Jiaotong University
2011-2024
Chengdu Medical College
2023
Nanjing University of Information Science and Technology
2023
Yellow River Institute of Hydraulic Research
2022
Hohai University
2021
East China Normal University
2020-2021
Chongqing Bureau of Geology and Minerals Exploration
2021
Ubiquitous Internet of Things (IoT) sensors in the smart city generate various urban utility sequential data, such as electricity and water usage records, which are defined multivariate time series (MUTS). Due to complex behavior human beings, MUTS contains more complicated relationships, go beyond general (TS). Specifically, multifaceted temporal couplings exist MUTS, including intra-/inter-TS, short-to-long term, evolving, polarized (positive/negative) relationships. Existing multisequence...
In the application of medium and long-term runoff forecasting, machine learning has some problems, such as high cost, limited computing difficulty in satisfying statistical data assumptions regions, leading to popularization hydrology industry. case a few data, it is one ways solve problem analyze characteristics consistency. This paper analyzes hypothesis periodicity mutation. Aiming at effect inconsistency on three representative models (multiple linear regression, random forest, back...
Air pollution seriously affects public health, while effective air quality prediction remains a challenging problem since the complex spatial-temporal couplings exist in multi-area monitoring data of city. Current approaches rarely consider relative geographical locations when capturing relations, instead latent inter-dependencies (i.e., implicit spatial relations) as replacement. However, such relations cannot necessarily reflect diffusion pollutants real world, and genuine location-related...
Autonomous umanned aerial vehicle (UAV) manipulation is necessary for the defense department to execute tactical missions given by commanders in future unmanned battlefield. A large amount of research has been devoted improving autonomous decision-making ability UAV an interactive environment, where finding optimal maneuvering policy became one key issues enabling intelligence UAV. In this paper, we propose a algorithm air-delivery based on deep reinforcement learning under guidance expert...
Ozone pollution is one of the most severe air quality issues in China that poses a serious threat to human health and ecosystems. During 2019-2021, maximum daily 8-hour average O3 concentrations eastern (110-122.5°E, 26-42°N) rest (ROC) show different decreasing patterns, with by 14.9 μg/m3, which much larger than 4.8 μg/m3 ROC. Here, based on atmospheric chemical transport model (GEOS-Chem) simulations machine learning (ML) (LightGBM) predictions, reasons for differences changes between ROC...
The prediction of express delivery sequence, i.e., modeling and estimating the volumes daily incoming outgoing parcels for delivery, is critical online business, logistics, positive customer experience, specifically resource allocation optimization promotional activity arrangement. A precise estimate consumer requests has to involve sequential factors such as shopping behaviors, weather conditions, events, business campaigns, their couplings. Despite that various methods have integrated...
Automatically detecting fake news as early possible becomes increasingly necessary. Conventional approaches of detection (FNED) verify news' veracity with a predefined and indiscriminate position, which depends on domain experience leads to unstable performance. More advanced methods address this problem proposed concept adaptive position (ADP), i.e. the where record can be concluded. Yet these either lack theoretical reliability or weaken complex dependencies among multi-aspect clues, thus...
Aimed at the problem of low reliability and significance level in multi-target threat assessment, based on advantage Vise Kriterijumska Optimizacija I Kompromisno Resenje (VIKOR) method which can take maximizing group effect minimize individual regret into account, setting reasonable decision-making mechanism coefficient, VIKOR is applied to assessment problems. The mathematic model established. And analyze influence changes coefficient final sorting results under conditions equal weight...
The traditional image restoration algorithm based on the sample block only considers simple structure information and will produce cumulative errors other issues in process of repair, which affected overall effect large part, to address this problem, an improved Criminisi is proposed, combines priority function new confidence item updating method. Simulation results show that can effectively overcome shortcomings excessive filling texture-rich area error derivative often appears.
The small-and medium-sized rivers are affected by the perennial dry flow. historical water level is scarce, which cannot meet number of training samples in neural network modeling. model's accuracy requirements when using traditional machine learning methods to predict level. An intelligent prediction model for based on small sample data proposed solve this problem. primary established Bayesian linear regression, and k-nearest neighbor algorithm introduced achieve first correction results....
Recently, content polluters post malicious information in Online Social Networks (OSNs), which is a more and serious problem that poses threat to the privacy information, account security, user experience, etc. They continuously simulate behaviors of legitimate accounts various ways, evade detection systems against them. In this paper, we focus on one kind polluter, namely collective polluter (hereinafter referred as CCP). Existing works either individual or require long periods data records...
Aiming at the influence of lens distortion to imaging measurement system, this paper puts forward an improved method for correction based on linear approximation. Firstly, solves radial coefficient linearly principle cross-ratio invariance under perspective projection. Then a higher precision straight line is obtained according orthogonal distance fitting. Finally, comprehensive index function built get via Gauss-Newton method. Simulation results show that average correcting error less than...
Hydrological big data was characterized by complexity and comprehensibility, There are massive association relationships to be mined in hydrological data. forecasting is primary flood prevention China, how use make accurate efficient prediction has become the study. This study starts with analysis of relationship, captures characteristics data, proposes shared weight Long Short- Term Memory(SWLSTM) reduce number optimized variables, shorten training time SWLSTM, improve accuracy Taking daily...
In recent years, the virtual-reality simulation technology has developed so fast which been widely used in military and people’s daily lives, etc. As essential element of three-dimensional scene, modeling trees becomes increasingly important. There are many methods for 3D such as 3DMAX modeling, bulletin board technology, crossing panel LOD technology. However, these not effective either bad real-time performance caused by large quantity data or unsatisfactory effect. Therefore, this paper...