UROP Project
Graph Neural Networks, Social Networks, Recommendation Systems, Scalable Machine Learning, Inductive Learning
Research Mentor: Dr. Yushun Dong, he/his/him
Department, College, Affiliation: Department of Computer Science, Arts and Sciences
Contact Email: yd24f@fsu.edu
Research Assistant Supervisor (if different from mentor):
Research Assistant Supervisor Email:
Faculty Collaborators:
Faculty Collaborators Email:
Department, College, Affiliation: Department of Computer Science, Arts and Sciences
Contact Email: yd24f@fsu.edu
Research Assistant Supervisor (if different from mentor):
Research Assistant Supervisor Email:
Faculty Collaborators:
Faculty Collaborators Email:
Looking for Research Assistants: No
Number of Research Assistants: 6
Relevant Majors: Computer Science, Electrical and Computer Engineering, Data Science, Applied Mathematics, Statistics
Project Location: On FSU Main Campus
Research Assistant Transportation Required: No, the project is remote Remote or In-person: Fully Remote
Approximate Weekly Hours: 5 - 10 hours,
Roundtable Times and Zoom Link: Thursday, Sept. 5th from 2PM - 2:30PM ET
Zoom link: https://fsu.zoom.us/j/7153751215
Number of Research Assistants: 6
Relevant Majors: Computer Science, Electrical and Computer Engineering, Data Science, Applied Mathematics, Statistics
Project Location: On FSU Main Campus
Research Assistant Transportation Required: No, the project is remote Remote or In-person: Fully Remote
Approximate Weekly Hours: 5 - 10 hours,
Roundtable Times and Zoom Link: Thursday, Sept. 5th from 2PM - 2:30PM ET
Zoom link: https://fsu.zoom.us/j/7153751215
Project Description
This project aims to develop novel graph learning frameworks to facilitate inductive and scalable recommendations on large-scale social networks. The research focuses on overcoming the limitations of existing Graph Neural Networks (GNNs) by designing a model that can be trained with limited computational costs and easily generalized to unseen social networks without retraining. The proposed framework will leverage both structural and positional encoding to achieve scalable and inductive recommendations, potentially improving the efficiency of recommendation systems on online social network platforms.Research Tasks: (1) Develop a scalable and inductive method for social network recommendation
(2) Design and implement a novel message-passing graph neural network model
(3) Implement and optimize the proposed graph learning framework
(4) Conduct offline evaluations using public and anonymous recommendation datasets
(5) Analyze and compare performance metrics such as NDCG and other industrial recommendation metrics with alternative models
Skills that research assistant(s) may need: (Recommended) Strong programming skills, particularly in Python
(Recommended) Experience with machine learning frameworks (e.g., PyTorch, TensorFlow)
(Recommended) Familiarity with recommendation systems and social network analysis
Mentoring Philosophy
As the principal investigator, I believe in fostering a collaborative and supportive research environment. Research assistants will have the opportunity to work closely with me and other team members, including PhD student Yushun Dong. We encourage creative thinking, rigorous analysis, and open communication. Research assistants will be given the opportunity to contribute to cutting-edge research in graph machine learning and recommendation systems, with the potential for co-authorship in research publications. We also value the development of practical skills through collaboration with our industry partners, bringing potential opportunities such as internships.Additional Information
The project builds upon the PI's strong research experience in graph machine learning, with opportunities to work on research paper submissions and research topics that are closely related to the listed one.Successful candidates will be able to continue working with the research group under a broader scope of collaborations leading to a track record of high-impact publications and industry collaborations.
If you are interested, please visit the site below for a toy research example. Please share your opinions with me to gain priority on working with me by reaching out to yd24f@fsu.edu.
https://yushundong.github.io/files/2024_toy_essay.pdf