James Murray Streitberg
Graph Fourier Analysis for Community Detection in Covert Networks
Much work has been done on detecting modular or community structures in covert networks by leveraging insights from Spectral Graph theory, particularly the spectral properties of the graph Laplacian in community
detection algorithms like Spectral Clustering (von Luxburg, 2007). This project extends prior work by conceptualising
connections between Spectral Graph theory and the Graph Fourier Transform (GFT). Low-frequency components correspond to smooth signals over the graph and may be associated with community structure (Ramakrishna et al., 2020; Shuman et al., 2013). The project will compare Fourier-driven methods with spectral methods, investigating where they relate or diverge through a literature review, and test GFT’s potential for large-scale anomaly detection in datasets such as ENRON or the Elliptic Bitcoin Dataset, comparing outputs and evaluating any correspondences between GFT and Spectral Graph theory.
James Murray Streitberg
RMIT University
James Streitberg is finishing his third year of his Bachelor of Applied Mathematics and Statistics with RMIT University and is looking forward to extending his studies in an Honours year. He has a strong interest in learning new mathematical fields and the unique problem-solving perspectives they offer. After completing a research project in spectral graph theory, James uncovered a passion for graph theory and network science, particularly their applications in community detection algorithms.
Outside of studies James is an avid reader of science fiction, fantasy, and mathematical history. He enjoys logic puzzles and strategy games, ideally accompanied by a good cup of coffee.
You may be interested in
Mikaela Westlake
Capturing the impact of patient variability in a novel cancer treatment using Bayesian inference