• Graduate Programs
  • Research
  • Browse our Courses
  • Events
    • Events Calendar
    • Events Archive
    • Summer School
      • Applied Public Policy Evaluation
      • Deep Learning
      • Development Economics
      • Economics of Blockchain and Digital Currencies
      • Economics of Climate Change
      • The Economics of Crime
      • Foundations of Machine Learning with Applications in Python
      • From Preference to Choice: The Economic Theory of Decision-Making
      • Inequalities in Health and Healthcare
      • Marketing Research with Purpose
      • Markets with Frictions
      • Modern Toolbox for Spatial and Functional Data
      • Sustainable Finance
      • Tuition Fees and Payment
      • Business Data Science Summer School Program
    • Tinbergen Institute Lectures
    • 2026 Tinbergen Institute Opening Conference
    • Annual Tinbergen Institute Conference
  • News
  • Summer School
    • Applied Public Policy Evaluation
    • Deep Learning
    • Development Economics
    • Economics of Blockchain and Digital Currencies
    • Economics of Climate Change
    • The Economics of Crime
    • Foundations of Machine Learning with Applications in Python
    • From Preference to Choice: The Economic Theory of Decision-Making
    • Inequalities in Health and Healthcare
    • Marketing Research with Purpose
    • Markets with Frictions
    • Modern Toolbox for Spatial and Functional Data
    • Sustainable Finance
    • Tuition Fees and Payment
  • Alumni
Home | Events Archive | Toward an Interpretable Criminal Footprint: Penalised Logistic Regression and Graph-Theoretic Features for Bitcoin Forensics
Research Master Pre-Defense

Toward an Interpretable Criminal Footprint: Penalised Logistic Regression and Graph-Theoretic Features for Bitcoin Forensics

  • Location
    Roeterseilandcampus RecE 4.03
    Amsterdam
  • Date and time

    July 07, 2026
    11:00 - 13:00

This thesis analyses a labelled Bitcoin transaction dataset widely used in illicit activity detection, yielding insights valuable for anti-money laundering (AML) and counter-terrorism financing (CTF) legislation. The study employs a penalised logistic regression framework alongside a theory-grounded selection of network centrality measures, a combination that could demonstrates strong out-of-sample predictive performance previously overlooked by the research community. Furthermore, this work identifies critical methodological flaws in the original dataset and contributes an improved, extended version as a valuable resource for future research.