Research Projects
Large Language Models + Graph Neural Networks
A method for predicting congressional voting results based on large language models and hypergraphs, combining GNN and LLM technologies for behavior prediction and information propagation analysis in social networks.
Research and Application of Image Privacy Protection Framework Based on Flow Model
This project proposes two image privacy protection schemes to address the security issues in data transmission and the balance between privacy and usability, and verifies their practicality in embedded systems.
Controlling the Development of Epidemic Diseases Using Trajectory Networks and Social Networks
This project proposes a method for controlling the development of epidemic diseases using trajectory networks and social networks, which can help policymakers determine which policies can more effectively mitigate the spread of the epidemic.
Modeling of Social Networks Based on Key Opinion Leaders (KOLs)
This project is based on the KOL user model of Internet social network platforms, using scientific methods to extract behavioral characteristic parameters of different types of users in large and complex networks.
Detection and Localization of Internal Malicious Attacks in Distributed Computing Networks
This project provides guidelines on how to design defense of multi-agent algorithms and improve the network's immunity to various attacks.
Trajectory Network Modeling and Analysis
This project performs user trajectory feature extraction and KOL modeling on the transportation network, which enables us to profile each user of the transportation network.
Distributed Blockchain Network Data Analysis and Research
This project conducts KOL modeling on this distributed social network, quantitatively analyzes the network involving smart contract transactions, understands the relationship between network clusters, and discovers key users.
5G Mobile Phone Signal Fingerprint Identification
This project performs fingerprint identification on 5G mobile phone signals, extracts the fingerprint feature information of the mobile phone signal, and identifies and detects the radiation source.
Research on Distributed Principal Component Analysis Algorithm
This project investigates the performance of different distributed PCA methods on different data sets, their performance, and their application in distributed data collection systems.