Top of page

Student/Staff Portal
Global Site Navigation

School of Science

Local Section Navigation
You are here: Main Content

Related Contents

Daniel Kam

Overview of Thesis

When conducting the initial triage or survey in a digital forensic examination, investigators rely on digital forensic string search tools which presuppose prior knowledge of both the evidence and the precise search terms required to retrieve it. The key challenges lie in addressing two fundamental questions: What does evidence look like? And how do we find it?

In response to these challenges, this research proposes a data mining and knowledge discovery (KDD) approach that leverages topic modelling, an unsupervised machine learning technique. The research begins by developing a forensic index search tool to enable the targeted exploration and exportation of the Digital Forensic Tool (DFT) search index. Three topic modelling algorithms are evaluated. The topic model pipeline includes preprocessing, training, post-processing, and evaluation. For each model, statistical and AI-driven techniques are applied to obtain interpretability ratings, coherence scores, and topic annotations.

This PhD research aims to enhance digital forensic investigators' ability to contextualize digital evidence and retrieve relevant information more effectively. The expected outcome is a proof-of-concept module for an existing DFT, designed to optimize keyword searches and improve the overall understanding of forensic data. The proposed module serves as a form of augmented intelligence, providing insights that empower investigators in their work.

Research Interests

  • Data Mining and Knowledge Discovery techniques applied to the Search Index of a Digital Forensic Tool
  • Textual Analytics for Forensic Applications
  • Augmented Intelligence

Scholarships and Awards

  • Joint Scholarship (Edith Cowan University / Cyber Security Research Cooperative) 2021 - 2024

Supervisors

Contact

Daniel Kam
PhD Student
School of Science
Skip to top of page