A5046856721- Human Mobility and Location-Based Analysis
- Traffic Prediction and Management Techniques
- Transportation Planning and Optimization
- Urban Transport and Accessibility
- Time Series Analysis and Forecasting
- Video Surveillance and Tracking Methods
- Data Management and Algorithms
- Data-Driven Disease Surveillance
- 2D Materials and Applications
- Anomaly Detection Techniques and Applications
- Transportation and Mobility Innovations
- Geographic Information Systems Studies
- Advanced Graph Neural Networks
- Opportunistic and Delay-Tolerant Networks
- Advanced Image and Video Retrieval Techniques
- Evacuation and Crowd Dynamics
- Robotics and Sensor-Based Localization
- COVID-19 epidemiological studies
- Target Tracking and Data Fusion in Sensor Networks
- Stock Market Forecasting Methods
- Data Visualization and Analytics
- Mobile Crowdsensing and Crowdsourcing
- Recommender Systems and Techniques
- Topological Materials and Phenomena
- Graphene research and applications
Jilin University
2024-2025
Beijing Institute of Technology
2014-2025
Capital University of Economics and Business
2025
Qingdao Agricultural University
2025
Ministry of Industry and Information Technology
2025
Southern University of Science and Technology
2019-2024
The University of Tokyo
2015-2024
Northwestern Polytechnical University
2024
Jilin Medical University
2024
Khon Kaen University
2024
Channel shear connectors are known as an appropriate alternative for common due to having a lower manufacturing cost and easier installation process. The behavior of channel is generally determined through conducting experiments. However, these experiments not only costly but also time-consuming. Moreover, the impact other parameters cannot be easily seen in connectors. This paper aims investigate application hybrid artificial neural network–particle swarm optimization (ANN-PSO) model...
This paper assesses the potentiality of certainty factor models (CF) for best suitable causative factors extraction landslide susceptibility mapping in Sado Island, Niigata Prefecture, Japan. To test applicability CF, a inventory map provided by National Research Institute Earth Science and Disaster Prevention (NIED) was split into two subsets: (i) 70% landslides to be used building CF based model; (ii) 30% validation purpose. A spatial database with fifteen then constructed processing ALOS...
With the rapid development of urbanization and public transportation system, number traffic accidents have significantly increased globally over past decades become a big problem for human society. Facing these possible unexpected accidents, understanding what causes accident early alarms some ones will play critical role on planning effective management. However, due to lack supported sensing data, research is very limited field updating risk in real-time. Therefore, this paper, we collect...
Traffic forecasting as a canonical task of multivariate time series has been significant research topic in AI community. To address the spatio-temporal heterogeneity and non-stationarity implied traffic stream, this study, we propose Spatio-Temporal Meta-Graph Learning novel Graph Structure mechanism on data. Specifically, implement idea into Convolutional Recurrent Network (MegaCRN) by plugging Learner powered Meta-Node Bank GCRN encoder-decoder. We conduct comprehensive evaluation two...
Nowadays, with the rapid development of IoT (Internet Things) and CPS (Cyber-Physical Systems) technologies, big spatiotemporal data are being generated from mobile phones, car navigation systems, traffic sensors. By leveraging state-of-the-art deep learning technologies on such data, urban prediction has drawn a lot attention in AI Intelligent Transportation System community. The problem can be uniformly modeled 3D tensor (T, N, C), where T denotes total time steps, N size spatial domain...
With the rapid development of Intelligent Transportation System (ITS), accurate traffic forecasting has emerged as a critical challenge. The key bottleneck lies in capturing intricate spatio-temporal patterns. In recent years, numerous neural networks with complicated architectures have been proposed to address this issue. However, advancements network encountered diminishing performance gains. study, we present novel component called adaptive embedding that can yield outstanding results...
Multivariate time-series (MTS) forecasting is a paramount and fundamental problem in many real-world applications. The core issue MTS how to effectively model complex spatial-temporal patterns. In this paper, we develop adaptive, interpretable scalable framework, which seeks individually each component of the We name framework SCNN, as an acronym <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</u> tructured...
To improve the accuracy of steel surface defect detection, an improved model multi-directional optimization based on YOLOv8 algorithm was proposed in this study. First, we innovate CSP Bottleneck with two convolutions (C2F) module by introducing deformable convolution (DCN) technology to enhance learning and expression ability complex texture irregular shape features. Secondly, advanced Bidirectional Feature Pyramid Network (BiFPN) structure is adopted realize weight distribution input...
The frequency and intensity of natural disasters has significantly increased over the past decades this trend is predicted to continue. Facing these possible unexpected disasters, accurately predicting human emergency behavior their mobility will become critical issue for planning effective humanitarian relief, disaster management, long-term societal reconstruction. In paper, we build up a large database (GPS records 1.6 million users one year) several different datasets capture analyze...
Human movements are difficult to predict, especially, when we consider rare behaviors that deviate from normal daily routines. By tracing the behavior of a person over long period, can model their routines and predict periodical behaviors, whereas such as participating in New Year's Eve countdown, hardly be predicted readily thus they have usually been treated outliers most existing studies. However, for scenarios emergency management or intelligent traffic regulation, more interested than...
Multi-variate time series (MTS) data is a ubiquitous class of abstraction in the real world. Any instance MTS generated from hybrid dynamical system with their specific dynamics normally unknown. The nature such result complex external impacts, which can be summarized as high-frequency and low-frequency temporal view, or global local if we take spatial view. These impacts also determine forthcoming development making them paramount to capture forecasting task. However, conventional methods...
Event crowd management has been a significant research topic with high social impact. When some big events happen such as an earthquake, typhoon, and national festival, becomes the first priority for governments (e.g. police) public service operators subway/bus operator) to protect people's safety or maintain operation of infrastructures. However, under event situations, human behavior will become very different from daily routines, which makes prediction dynamics at highly challenging,...
The bathymetry of nearshore coastal environments and lakes is constantly reworking because the change in patterns energy dispersal related sediment transport pathways. Therefore, updated accurate bathymetric models are a crucial component providing necessary information for scientific, managerial, geographical studies. Recent advances satellite technology revolutionized acquisition profiles, offering new vistas mapping. This contribution analyzed suitability Sentinel-2 Landsat-8 images...
The photovoltaic (PV) industry boom and increased PV applications call for better planning based on accurate updated data the installed capacity. Compared with manual statistical approach, which is often time-consuming labor-intensive, using satellite/aerial images to estimate existing capacity offers a new method cost-effective data-consistent features. Previous studies investigated feasibility of segmenting panels from involving machine learning technologies. However, due particular...
Abstract Understanding Mott insulators and charge density waves (CDW) is critical for both fundamental physics future device applications. However, the relationship between these two phenomena remains unclear, particularly in systems close to two-dimensional (2D) limit. In this study, we utilize scanning tunneling microscopy/spectroscopy investigate monolayer 1T-NbSe 2 elucidate energy of upper Hubbard band (UHB), reveal that spin-polarized UHB spatially distributed away from dz orbital at...
Abstract Chirality is essential for various phenomena in life and matter. However, chirality its switching electronic superlattices, such as charge density wave (CDW) remain elusive. In this study, we characterize the with atom-resolution imaging a single-layer NbSe 2 CDW superlattice by technique of scanning tunneling microscopy. The atomic arrangement found continuous intact although switched. Several intermediate states are tracked time-resolved imaging, revealing fast dynamic transition....
In recent years, the use of WiFi fingerprints for indoor positioning has grown in popularity, largely due to widespread availability and proliferation mobile communication devices. However, many existing methods constructing fingerprint data sets rely on labor-intensive time-consuming processes collecting large amounts data. Additionally, these often focus ideal laboratory environments, rather than considering practical challenges multifloor buildings. To address issues, we present a novel...
Spatiotemporal forecasting techniques are significant for various domains such as transportation, energy, and weather. Accurate prediction of spatiotemporal series remains challenging due to the complex heterogeneity. In particular, current end-to-end models limited by input length thus often fall into mirage, i.e., similar time followed dissimilar future values vice versa. To address these problems, we propose a novel self-supervised pre-training framework Spatial-Temporal-Decoupled Masked...