A5012555900- Face and Expression Recognition
- Sparse and Compressive Sensing Techniques
- Remote-Sensing Image Classification
- EEG and Brain-Computer Interfaces
- Rough Sets and Fuzzy Logic
- Text and Document Classification Technologies
- Gaze Tracking and Assistive Technology
- Advanced Clustering Algorithms Research
- Image Retrieval and Classification Techniques
- Neural Networks and Applications
- Machine Learning and ELM
- Quantum Computing Algorithms and Architecture
- Domain Adaptation and Few-Shot Learning
- Computational Drug Discovery Methods
- Machine Learning in Bioinformatics
- Medical Image Segmentation Techniques
- Advanced Computing and Algorithms
- Quantum Information and Cryptography
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
- Environmental Toxicology and Ecotoxicology
- Advanced Graph Neural Networks
- Obstructive Sleep Apnea Research
- Data Mining Algorithms and Applications
- Toxic Organic Pollutants Impact
Tongji University
2006-2025
Sun Yat-sen Memorial Hospital
2025
Sun Yat-sen University
2025
Shanghai Maritime University
2015-2024
Nanjing University of Science and Technology
2024
Hunan Normal University
2024
Hunan Provincial People's Hospital
2024
Yili Normal University
2024
Fudan University
2023
Zhongshan Hospital
2023
Drug-drug interaction (DDI) defines a situation in which one drug affects the activity of another when both are administered together. DDI is common cause adverse reactions and sometimes also leads to improved therapeutic effects. Therefore, it great interest discover novel DDIs according their molecular properties mechanisms robust rigorous way. This paper attempts predict effective using following properties: (1) chemical between drugs; (2) protein interactions targets (3) target...
This paper concerns the problem of network embedding (NE), whose aim is to learn low-dimensional representations for nodes in networks. Such dense vector offer great promises many analysis problems. However, existing NE approaches are still faced with challenges posed by characteristics complex networks real-world applications. First, associated rich content information, previous methods tend separated and structure each node, which requires a post-processing combination. The empirical...
Spectral-type subspace clustering algorithms have shown excellent performance in many applications. The existing spectral-type either focus on designing constraints for the reconstruction coefficient matrix or feature extraction methods finding latent features of original data samples. In this paper, inspired by graph convolutional networks, we use convolution technique to develop a method and constraint simultaneously. And graph-convolutional operator is updated iteratively adaptively our...
In the original belief function (BF) theory, a precise-valued structure has been widely used to represent uncertain information. However, this mentioned is difficult effectively measure specific hesitant situation, especially when decision makers have set of possible values for assignments focal elements. order model nature behavior people make under uncertainty, we propose fuzzy (HFBS) that based on BF theory and recent theory. We also present novel rule combination HFBS evaluated in...
While crystal structure prediction (CSP) remains a longstanding challenge, we introduce ParetoCSP, novel algorithm for CSP, which combines multi-objective genetic (GA) with neural network inter-atomic potential model to find energetically optimal structures given chemical compositions. We enhance the updated GA (NSGA-III) by incorporating genotypic age as an independent optimization criterion and employ M3GNet universal guide search. Compared GN-OA, state-of-the-art potential-based CSP...
Multiview subspace clustering (MVSC) aims to integrate complementary information from different views accurately reveal the structure of a multiview dataset. Traditional MVSC methods often emphasize aggregation samples within same subspace, while neglecting separation across subspaces. In this article, we incorporate contrastive learning techniques into framework, developing data self-representation module, regularizer for reconstruction coefficient matrix in each view, and alignment term...
The study aims to develop an automatic sleep scoring method by fusing different polysomnography (PSG) signals and further investigate PSG signals' contribution the result. Eight combinations of four modalities signals, namely electroencephalogram (EEG), electrooculogram (EOG), electromyogram (EMG), electrocardiogram (ECG) were considered find optimal fusion signals. A total 232 features, covering statistical characters, frequency time-frequency fractal entropy characters nonlinear derived...
AbstractGiven the widespread use as flame retardants, plasticizers, and organophosphate esters (OPEs) received increasing scientific interests on their occurrence ecotoxicological research progress. This review comprehensively conducted bibliometric analysis surveyed OPEs in aquatic ecosystem (water, sediment, organisms) human-related (drinking water sewage) over past decade to unraveling knowledge gaps. The concentrations water, sediment were at range of not detected (n.d.) or several...
This paper presents a novel method to classify human facial movement based on multi-channel forehead bio-signals. Five face movements form three regions: forehead, eye and jaw are selected classified in back propagation artificial neural networks (BPANN) by using combination of transient steady features from EMG EOG waveforms. The identified subsequently employed generate five control commands for controlling simulated intelligent wheelchair. A human-machine interface (HMI) is designed map...
Due to the corruptions or noises that existed in real-world data sets, affinity graphs constructed by classical spectral clustering-based subspace clustering algorithms may not be able reveal intrinsic structures of sets faithfully. In this article, we reconsidered reconstruction problem and proposed idea "relation reconstruction." We pointed out a sample could represented neighborhood relation computed between its neighbors itself. The indicate true membership corresponding original...