Document Type

Dissertation

Date of Award

8-31-2025

Degree Name

Doctor of Philosophy in Computing Sciences - (Ph.D.)

Department

Computer Science

First Advisor

David A. Bader

Second Advisor

Baruch Schieber

Third Advisor

Ioannis Koutis

Fourth Advisor

Senjuti Basu Roy

Fifth Advisor

Hang Liu

Abstract

Graph algorithms are essential analytical tools with applications spanning cybersecurity, biology, social media, and increasingly, financial technology (FinTech). The complex and interconnected nature of financial data, particularly in cryptocurrency networks, presents unique opportunities for graph-based analysis in fraud detection and anomaly identification.

This dissertation presents the design and implementation of scalable graph algorithms tailored for large-scale networks, with particular emphasis on FinTech applications. The primary contributions include: (1) novel cover-edge based triangle counting algorithms that significantly reduce computational overhead through breadth-first search preprocessing, achieving substantial speedups over traditional methods and dramatic communication reduction in distributed settings, (2) optimized parallel implementations of community detection algorithms, including Label Propagation, Louvain, and Leiden, that leverage system-level parallelism to achieve performance improvements while maintaining high-quality community structures, (3) practical applications of these algorithms to cryptocurrency transaction analysis for fraud detection, demonstrating how structural graph patterns can effectively identify illicit activities in financial networks.

Some algorithms were implemented within the Arachne framework, a high-performance graph analytics platform built on Chapel that provides a Python interface for accessibility while maintaining parallel performance. Experimental evaluation on both synthetic and real-world datasets validates the efficiency and scalability of the proposed methods. The fraud detection case studies demonstrate that fraudulent nodes tend to cluster together within detected communities, suggesting that community structure can serve as a valuable signal for identifying potentially suspicious activity in cryptocurrency networks.

This work bridges the gap between theoretical graph algorithm development and practical FinTech applications, providing efficient tools for analyzing massive financial networks while maintaining interpretability for forensic analysis and regulatory compliance.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.