In my research, I focus on the development and application of graph theory algorithms, particularly paraclique, for life science data analysis. This method, which centers around the critical "glom term" hyperparameter, is adept in clustering, outlier detection, and link prediction. My key contribution lies in enhancing the efficiency of identifying dense subgraphs in k-partite graphs, streamlining this intricate process by pruning extraneous search paths. Additionally, I am exploring graph neural networks. My work showcases a deep understanding of complex algorithms and a substantial impact in machine learning, data analysis, and algorithm optimization.
Determine the associations of the circulating microbiome with various clinical characteristics in ESRD patients
University of Tennessee, Knoxville, TN 01/2023~present
Disease Subtyping by Graph Theory Algorithms with Comprehensive Feature selection
University of Tennessee, Knoxville, TN 08/2023~present
Gene product-Disease-Drug Link Prediction Using Tripartite Graph
University of Tennessee, Knoxville, TN 08/2020~05/2022
Internships
Ph.D. Software Engineer Intern
Google, Mountain View, California, USA 05/2023-08/2023
Smart display anomaly detection and analysis by data mining
Data Scientist Intern
Amazon, Sunnyvale, California, USA 05/2022-08/2022
Building an ensemble system of multiple Machine Learning techniques
Working with big data and, building solutions to challenging problems that directly impact the company's bottom-line
Data Mining and Analysis Engineer Internship
S.F.Express Co.LTF, Shenzhen, China 05/2017-07/2017