A5088111638- Multimodal Machine Learning Applications
- Groundwater and Isotope Geochemistry
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Domain Adaptation and Few-Shot Learning
- Generative Adversarial Networks and Image Synthesis
- Groundwater flow and contamination studies
- Advanced Image Processing Techniques
- Hydrocarbon exploration and reservoir analysis
- Face recognition and analysis
- Hydrology and Watershed Management Studies
- Geological formations and processes
- Image and Signal Denoising Methods
- Maritime Transport Emissions and Efficiency
- Advanced Neural Network Applications
- Digital Media Forensic Detection
- Advanced Vision and Imaging
- Landslides and related hazards
- Anomaly Detection Techniques and Applications
- Advanced Chemical Sensor Technologies
- Topic Modeling
- Image Retrieval and Classification Techniques
- Groundwater and Watershed Analysis
- Geological and Geophysical Studies
- Advanced Steganography and Watermarking Techniques
Linyi University
2024-2025
Wuhan University of Technology
2019-2024
Institute of Genetics and Developmental Biology
2017-2024
Center for Agricultural Resources Research
2017-2024
Chang'an University
2018-2024
Foshan University
2024
First Automotive Works (China)
2024
Sanya University
2022-2023
Central China Normal University
2022-2023
Sinopec (China)
2014-2023
From AlphaGo to ChatGPT, the field of AI has launched a series remarkable achievements in recent years. Analyzing, comparing, and summarizing these at paradigm level is important for future innovation, but not received sufficient attention. In this paper, we give an overview perspective on machine learning paradigms. First, propose taxonomy with three levels seven dimensions from knowledge perspective. Accordingly, basic twelve extended paradigms, such as Ensemble Learning, Transfer etc.,...
Credit card transaction fraud costs billions of dollars to issuers every year. Besides, the credit dataset is very skewed, there are much fewer samples frauds than legitimate transactions. Due data security and privacy, different banks usually not allowed share their datasets. These problems make traditional model difficult learn patterns also detect them. In this paper, we introduce a novel framework termed as federated meta-learning for detection. Different from technologies trained with...
Depression is a critical problem in modern society that affects an estimated 350 million people worldwide, causing feelings of sadness and lack interest pleasure. Emotional disorders are gaining closely entwined with depression, because one contributes to understanding the other. Despite achievements two separate tasks emotion recognition depression detection, there has not been much prior effort build unified model can connect these different modalities, including multimedia (text, audio,...
Faults play a crucial role in shaping the formation and damage patterns of landslides mountainous region, particularly Qinling-Daba (Qinba) area China. On 6 October 2022, following 4-day rainfall event totaling 221.5 mm, landslide occurred Hanwang Town, Shaanxi Province. The left boundary coincided with fault, which influence movement development landslide. To further understand quantified process mechanism landslide, comprehensive study was conducted, incorporating field investigations,...
As the size and source of network traffic increase, so does challenge monitoring analyzing traffic. The challenging problems classifying encrypted are imbalanced property data, generalization on an unseen dataset, overly dependent data size. In this paper, we propose application a meta-learning approach to address these in classification, named Flow-Based Relation Network (RBRN). RBRN is end-to-end classification model that learns representative features from raw flows then classifies them...
Understanding the interaction of surface water and groundwater affected by anthropogenic activities is great importance for resource quality management. The Xiong’an New Area, located in North China Plain, has been designated a new building area China’s government. Groundwater over pumped artificial was transferred to meet supply this region. Therefore, natural greatly changed there complex impact from activities. In study, we used chemical ions stable isotopes δ2H δ18O assess Area. We...
In this article, we propose a key secret-sharing technology based on generative adversarial networks (GANs) to address three major problems in the blockchain: 1) low security; 2) hard recovery of lost keys; and 3) communication efficiency. our scheme, proposed network plays role dealer treats process as classification issue. The idea is view secret an image during process. If user's private text, can covert text into called original image. Specifically, first divide subimages by...
Facial expression synthesis has gained increasing attention with the development of Generative Adversarial Networks (GANs). However, it is still very challenging to generate high-quality facial expressions since overlapping and blur commonly appear in generated images especially regions rich features such as eye mouth. Generally, existing methods mainly consider face a whole without paying specific characteristics expressions. In fact, according physiological psychological research,...
The geological hazards of landslides in Hanwang Town, Ziyang County, Ankang City, Shaanxi Province, have emerged, yet the current understanding spatial distribution characteristics and influencing factors this area remains unclear. Combining results remote sensing interpretation field investigation, seven factors, namely, elevation, slope direction, gradient, distance from rivers, faults, engineering geologic lithology, roads, are selected for study, each factor influence resolution Digital...
When stress waves propagate in rock slopes, they are influenced by various factors such as structural plane characteristics (e.g., stiffness, number, spacing, thickness), sawtooth angle of planes), filling materials (fully filled, partially and different materials), wave impedance properties, leading to phenomena amplitude attenuation, signal delay, velocity reduction. In this paper, using the one-dimensional Hopkinson bar model an example, a numerical method is employed analyze propagation...