A5103274467- EEG and Brain-Computer Interfaces
- Advanced Graph Neural Networks
- Complex Network Analysis Techniques
- Functional Brain Connectivity Studies
- Network Security and Intrusion Detection
- Traditional Chinese Medicine Studies
- Digital Media Forensic Detection
- Forensic Fingerprint Detection Methods
- Currency Recognition and Detection
Nanjing University of Information Science and Technology
2024
Carnegie Mellon University
2022-2023
Anomaly detection in graphs has attracted considerable interests both academia and industry due to its wide applications numerous domains ranging from finance biology. Meanwhile, graph neural networks (GNNs) is emerging as a powerful tool for modeling data. A natural fundamental question that arises here is: can abnormality be detected by networks? In this paper, we aim answer question, which nontrivial. As many existing works have explored, seen filters signals, with the favor of low...
High-performance automated detection methods for epilepsy play a crucial role in clinical diagnostic support. To address the challenge of effectively extracting features from epileptic EEG signals, characterized by strong spontaneity and complexity, novel feature extraction approach based on Window Kullback-Leibler Divergence (WKLD) is proposed, coupled with discrete wavelet analysis signal extraction. Then, Residual Multidimensional Taylor Network (ResMTN) classifier applied state...
Network embedding, a graph representation learning method illustrating network topology by mapping nodes into lower-dimension vectors, is challenging to accommodate the ever-changing dynamic graphs in practice. Existing research mainly based on node-by-node embedding modifications, which falls dilemma of efficient calculation and accuracy. Observing that dimensions are usually much smaller than number nodes, we break this with novel paradigm rotates scales axes space instead update....
Numerous computer vision algorithms have been designed to help detect the values of paper banknotes for blind and visually impaired individuals. Previous relied mostly on traditional methods that are less efficient, accurate, transferable than a convolutional neural network. So in our study, we used networks perform banknote image classification. We trained network with 1,000 Thai images. Our could classify real-life study an accuracy 100%; however, also found out model is vulnerable...