A5101735891- Complex Network Analysis Techniques
- Remote-Sensing Image Classification
- Face and Expression Recognition
- Remote Sensing and Land Use
- Advanced Computing and Algorithms
- Metaheuristic Optimization Algorithms Research
- Advanced Image Fusion Techniques
- Advanced Graph Neural Networks
- Opinion Dynamics and Social Influence
- Text and Document Classification Technologies
- Image Retrieval and Classification Techniques
- Artificial Immune Systems Applications
- Sparse and Compressive Sensing Techniques
- Advanced Clustering Algorithms Research
- Neural Networks and Applications
- Bioinformatics and Genomic Networks
- Advanced SAR Imaging Techniques
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Machine Learning and ELM
- Photoacoustic and Ultrasonic Imaging
- Evolutionary Algorithms and Applications
- stochastic dynamics and bifurcation
- Data Visualization and Analytics
- Synthetic Aperture Radar (SAR) Applications and Techniques
Xidian University
2016-2024
Nanjing Foreign Language School
2023
Longdong University
2023
Nanjing University of Aeronautics and Astronautics
2023
Southwest University of Science and Technology
2021
PLA Army Engineering University
2013-2017
Space Engineering University
2017
Multilabel feature selection solves the dimension distress of high-dimensional multilabel data by selecting optimal subset features. Noisy and incomplete labels raw hinder acquisition label-guided information. In existing approaches, mapping label space to a low-dimensional latent semantic decomposition mitigate noise is considered an effective strategy. However, decomposed contains redundant information, which misleads capture potential relevance. To eliminate effect information on...
Partial multi-label learning (PML) is defined as the construction of robust classification models from a training set where all instances are correlated with corresponding group candidate labels that only partially accurate. Existing PML approaches have attempted to elicit reliable by parsing guideline information labels. However, other side in describes what sample does not contain largely ignored. Moreover, existing focus on distinguishing noise and lack effective use information. To this...
Existing fusion-based local community detection algorithms have achieved good results. However, when assigning a node to community, similarity functions are sometimes used, which only use information, while ignoring connection information within the community. These fail find influential nodes, eventually leads failure complete To address these problems, new algorithm is proposed in this article. Two strategies, of strong fusion followed by weak fusion, used alternately fuse nodes. Compared...
Recently, most convolutional neural network-based methods use kernels of fixed size to extract features, which ignore the inherent spatial structure information ground objects and lose details. In addition, rough first-order statistics is not enough capture subtle differences between different categories non local context information. To address these issues, a hyperspectral image (HSI) classification method based on multi-scale cross-branch response second-order channel attention (MCRSCA)...
Pre-trained large language models demonstrate potential in extracting information from DNA sequences, yet adapting to a variety of tasks and data modalities remains challenge. To address this, we propose DNAGPT, generalized pre-training model trained on over 200 billion base pairs all mammals. By enhancing the classic GPT with binary classification task (DNA sequence order), numerical regression (guanine-cytosine content prediction), comprehensive token language, DNAGPT can handle versatile...
In the information age of big data, and increasingly large complex networks, there is a growing challenge understanding how best to restrain spread harmful information, for example, computer virus. Establishing models propagation node immunity are important parts this problem. article, dynamic immune model, based on community structure threshold (NICT), proposed. First, network model established, which regards nodes carrying as new in network. The method establishing edge between original...