A5079448948- Face and Expression Recognition
- Hydraulic Fracturing and Reservoir Analysis
- Rough Sets and Fuzzy Logic
- Hydrocarbon exploration and reservoir analysis
- Data Mining Algorithms and Applications
- Video Surveillance and Tracking Methods
- Reservoir Engineering and Simulation Methods
- Sparse and Compressive Sensing Techniques
- Remote Sensing and Land Use
- Advanced Algorithms and Applications
- Geological Studies and Exploration
- Advanced Neural Network Applications
- Drilling and Well Engineering
- Higher Education and Teaching Methods
- Oil and Gas Production Techniques
- Machine Learning and ELM
- Seismic Imaging and Inversion Techniques
- Advanced Manufacturing and Logistics Optimization
- Domain Adaptation and Few-Shot Learning
- Industrial Technology and Control Systems
- Synthesis and Biological Evaluation
- Text and Document Classification Technologies
- Metaheuristic Optimization Algorithms Research
- Data Management and Algorithms
- Tensor decomposition and applications
Research Institute of Petroleum Exploration and Development
2016-2025
Anhui Institute of Information Technology
2024-2025
Northwestern Polytechnical University
2025
Tianjin Normal University
2009-2024
Yangzhou University
2024
Sichuan University of Science and Engineering
2024
Zhejiang University
2019-2024
Shandong University
2024
Jinan University
2024
Hefei University of Technology
2022-2024
Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, which future object can be predicted based on previous scenes, we propose a grid representation method that retain fine-scale structure network. Network-wide speeds are converted into series static images input novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for forecasting. The...
The low-rank tensor representation (LRTR) has become an emerging research direction to boost the multi-view clustering performance. This is because LRTR utilizes not only pairwise relation between data points, but also view of multiple views. However, there one significant challenge: uses nuclear norm as convex approximation provides a biased estimation rank function. To address this limitation, we propose generalized nonconvex (GNLTA) for subspace clustering. Instead correlation, GNLTA...
Feature selection approaches based on mutual information can be roughly categorized into two groups. The first group minimizes the redundancy of features between each other. second maximizes new classification providing for selected subset. A critical issue is that large does not signify little redundancy, and vice versa. Features with but high may by group, low relevance classes highly scored group. Existing fail to balance importance both terms. As such, a term denoted as Independent...
Based on a large amount of core analysis data in eastern Pre-Caspian Basin, the relationship between permeability and porosity its influencing factors are studied. The sedimentary environments Carboniferous System Basin include open platform, restricted platform evaporate platform. For dolomite reservoirs there three main combination patterns pores, namely, inter-crystalline solution micro-pores, intra-crystalline among which first highest permeability. limestone reservoirs, pores...
Some existing low-rank approximation approaches either need to predefine the rank values (such as matrix/tensor factorization-based methods) or fail consider local information of data (e.g., spatial spectral smooth structure). To overcome these drawbacks, this paper proposes a new model called tensor nuclear norm-based with total variation regularization (TLR-TV) for color and multispectral image denoising. TLR-TV uses norm encode global prior preserve spatial-spectral continuity in unified...
Carboniferous carbonate reservoirs at the eastern edge of Pre-Caspian Basin have undergone complex sedimentation, diagenesis and tectonism processes, developed various reservoir space types pores, cavities fractures with complicated combination patterns which create intricate pore-throats structure. The pore-throat structure leads to porosity-permeability relationship, bringing great challenges for classification evaluation efficient development. Based on comprehensive analysis cores, thin...
Cancer classification is the critical basis for patient-tailored therapy. Conventional histological analysis tends to be unreliable because different tumors may have similar appearance. The advances in microarray technology make individualized therapy possible. Various machine learning methods can employed classify cancer tissue samples based on data. However, few elegantly adopted generating accurate and reliable as well biologically interpretable rules. In this paper, we introduce an...
This paper addresses the multi-view subspace clustering problem and proposes self-paced enhanced low-rank tensor kernelized (SETKMC) method, which is based on two motivations: (1) singular values of representations multiple instances should be treated differently. The reasons are that larger usually quantify major information less penalized; samples with different degrees noise may have various reliability for clustering. (2) many existing methods cause degraded performance when features...
Carbonate reservoirs have various types of reservoir spaces and complex pore structures, so the evaluation microscopic structures is great significance to favorable identification. In order accurately characterize micro-pore structure carbonate reservoir, this paper uses NMR experiment, high-pressure mercury injection, logging data establish a conversion model between T2 spectrum capillary pressure curve by piecewise power function method. The nuclear magnetic mainly bimodal, with small...
Too many input features in applications may lead to over-fitting and reduce the performance of learning algorithm. Moreover, most cases, each feature containing different information content has effects on prediction target. Therefore, a selection method for calculating importance feature, called WKNNGAFS, is proposed this paper. In method, genetic algorithm (GA) adopted search optimal weight vector, value ith component which corresponds contribution degree classification from global...