A5101800298- Radioactivity and Radon Measurements
- Earthquake Detection and Analysis
- Geophysical Methods and Applications
- Web Data Mining and Analysis
- Expert finding and Q&A systems
- Advanced Text Analysis Techniques
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Hydrocarbon exploration and reservoir analysis
- Radioactive element chemistry and processing
- Topic Modeling
- Image and Signal Denoising Methods
- Rock Mechanics and Modeling
- Medical Research and Treatments
- Geoscience and Mining Technology
- Seismology and Earthquake Studies
- Text and Document Classification Technologies
- Microwave Imaging and Scattering Analysis
- Geophysical and Geoelectrical Methods
- Groundwater and Isotope Geochemistry
- Advanced Graph Neural Networks
- Geochemistry and Geologic Mapping
- Safety and Risk Management
- Advanced Decision-Making Techniques
- Groundwater flow and contamination studies
East China University of Technology
2020-2025
Institute of Computing Technology
2021-2024
Northwest University
2020
State Key Laboratory of Continental Dynamics
2020
The edge detection method based on the magnetic gradient tensor data plays an important role in exploration because it is free from geomagnetic interference and contains more abundant information. This paper proposes a new anomaly using components. model established to compare with other methods, such as directional total horizontal derivative (THDz), analytical signal (AS), tilt angle, theta map, so on, under conditions of vertical magnetization, oblique noise interference. Through study...
The sandstone-type uranium deposits located within the Songliao Basin of China, which are noted for their significant reserves and low mining costs, have become a primary focus exploration in country. Accurate detection such anomalies is essential assessment resources. Traditional logging identification methods face challenges, including accuracy, slow recognition speeds, limited generalization capabilities. With advancements technology, artificial intelligence has introduced novel research...
In this study, long-term and continuous monitoring of atmospheric radon concentration, temperature, air pressure, humidity was conducted at China Jinping Underground Laboratory. The impacts humidity, pressure on concentration in the experimental environment were specifically examined, along with potential interactions among these factors. Moreover, Radon data denoised using Singular Spectrum Analysis (SSA) to reveal factors that might influence changes concentration. possible mechanisms by...
Trajectory data association algorithm is an important part of multi-target tracking method. Traditional methods require priori information such as target motion model and clutter density to conduct association, the accuracy low in complex scenes trajectory accompaniments crosses. Based on sensitivity BiGRU time series, this paper proposes improved network: BiGRU-DA. In paper, output gate structure added basis existing model, which can optimize results extract influence timing sequence...
Abstract As a kind of time series data in the field transportation, trajectory has complex multi-dimensional information. How to fully excavate information and analyse its dynamic pattern is an important problem be solved classification series. In response above problems, this paper improves existing self-attention mechanism, proposes algorithm based on convolutional mechanism. According characteristics data, mechanism first extracts fragments obtain flight status characteristics, then...
Cross-modal retrieval has been widely used in the Vision-Language field and achieved many results, but there is a lack of research trajectory-text field. At same time, current popular cross-modal models not only fine-grained semantic alignment between different modalities, also ignore influence grammatical structure text on effect. To solve above problems, this paper proposes dual-stream trajectory model combined with graph neural network, combining local global two interaction methods: (1)...
Event extraction is a key research direction in the field of information extraction. In order to improve effect event and solve problem that general method cannot make full use text feature information, an integrating trigger word features proposed. By building remote thesaurus, we can provide additional for type classification model, enhance ability discovering words. Then arguments model integrates distance representation learning ability. Finally, connecting series complete Experiments...
In geogas measurement, in order to further find out the influence of exposed fracture zone on uranium element migrating with groundwater, numerical analysis was carried by finite method.