Zhen Guo
I am a Ph.D. new grad in Computer Science at Virginia Tech with 6-8 years of degree-seeking and intern experience. I am seeking NLP/ML Researcher or Engineer positions. I worked on my dissertation in tClab, advised by Professor Jin-Hee Cho. I had my MS in Computer Science, MS in Biological Sciences, and Advanced Certificate in Financial Econometrics and Data Analytics in LIDM lab at Fordham University, advised by Clavius Distinguished Professor Frank Hsu.
Research Interests
My research expertise is developing Artificial Intelligence (AI)-driven social systems to understand and combat online social deception attacks, such as phishing, fake news, false information, and cyber-grooming. My technical skills cover natural language processing (NLP) solutions by natural language generation (NLG), such as domain-specific chatbots (GPT, BERT, and T5); and by natural language understanding (NLU), such as intent prediction of fake news (RNN and LSTM) from tweets and posts, and users’ social-psychological feature engineering and text mining (LIWC). Furthermore, I fine-tune the Large Language Model (LLM) and the RNN by Deep Reinforcement Learning (DRL) models such as Actor-Critic (AC), Advantage Actor-Critic (A2C), Proximal Policy Optimization (PPO) in the NLP domain. In addition, I developed robust friend recommendation social systems and dynamic opinion propagation systems based on social capital, social and behavioral theories, opinion dynamics, information deception theory, and networked game theory.
Featured Projects
Local Uncertainty Reward in DRL System for Text Classification
A series of policy gradient deep reinforcement learning intent classifiers with a new multi-dimentional uncertainty reward function.
Generative Chatbot System SERI to Defend Online Predators
A generative chatbot framework of a perpetrator chatbot in cybergrooming for increasing youths sensitivity and awareness.
Text Mining Features for Cybergrooming Victims Vulnerability
A case study of text mining-based social-psychological vulnerability analysis of potential victims to cybergrooming.
Social Information Intent Classifier
A deep reinforcement learning REINFORCE intent classifier of social media information with a reward of two uncertainty metrics.
Uncertain Opinion Game
A social game system of uncertainty-based opinion interactions to mitigate network polarization of disinformation propagation.
Social System SAFER
A social capital-based friend recommendation system to defend against phishing attacks.
Opinion Dynamics Game
A three-player game framework of users' five Subjective Logic opinion models to defend disinformation propagation.
Online Social Deception Survey
An extensive survey of online social deception (OSD) attacks types and defense mechanisms of prevention, detection, and response (mitigation).
Uncertainty Reasoning Survey
A survey of uncertainty reasoning and quantification in belief theories and their applicability in modern deep learning.
Selected Publications
*Please see my full list of publications on my Google Scholar page.
Zhen Guo, Pei Wang, Jin-Hee Cho, Lifu Huang
[UDM Workshop, AAAI 2023] Uncertainty-Aware Reward-based Deep Reinforcement Learning for Intent Analysis of Social Media Information.
Zhen Guo, Qi Zhang, Xinwei An, Qisheng Zhang, Audun Jøsang, Lance Kaplan, Feng Chen, Dong Hyun Jeong, Jin-Hee Cho
Zhen Guo, Jin-Hee Cho, Chang-Tien Lu
Zhen Guo, Jaber Valinejad, Jin-Hee Cho
Zhen Guo, Jin-Hee Cho, Ing-Ray Chen, Srijan Sengupta, Michin Hong, Tanushree Mitra
Pei Wang, Zhen Guo, Lifu Huang, Jin-Hee Cho
My Life
Get in touch at [zguo] at [vt] dot [edu]
