View Proposal

---

Proposer

Timothy Yap

Title

Characterizing Cryptocurrency User Behaviour Through Transaction Graph Analytics

Goal

To investigate whether transaction graph characteristics can be used to identify and characterize behavioural patterns of cryptocurrency users.

Description

Blockchain networks generate large volumes of publicly accessible transaction data that can be represented as transaction graphs. These graphs provide a rich source of information for understanding how users interact with cryptocurrency networks and how different types of participants exhibit distinct behavioural patterns. This project aims to analyze cryptocurrency transaction graphs and investigate whether graph-based features can be used to characterize different classes of blockchain users. The study will extract structural and temporal features from transaction networks and evaluate their effectiveness in identifying behavioural patterns among wallets and addresses. Potential features include transaction frequency, network centrality measures, connectivity characteristics, clustering coefficients, and activity patterns over time. Machine learning models may be applied to analyze relationships between graph features and user behaviour categories. The project will contribute to a better understanding of blockchain ecosystems and demonstrate how graph analytics can be applied to cryptocurrency transaction data for behavioural analysis. Research Questions: Can graph-based features distinguish different categories of cryptocurrency users? Which graph characteristics are most useful for behavioural analysis? How do transaction patterns evolve over time? Can machine learning models effectively classify behavioural patterns from transaction graph features? Potential Evaluation: Feature importance analysis Classification performance comparison Graph structure analysis Temporal behaviour analysis Visualization of transaction networks

Resources

Python, NetworkX, Graph Analytics Libraries, Scikit-learn, Public Cryptocurrency Transaction Datasets, Bitcoin or Ethereum Blockchain Data Sources

Background

Python Programming, Data Analytics, Basic Machine Learning, Graph Theory Fundamentals, Blockchain Fundamentals

Url

External Link

Difficulty Level

High

Ethical Approval

Full

Number Of Students

1

Supervisor

Timothy Yap

Keywords

blockchain analytics, transaction graphs, cryptocurrency, graph analytics, machine learning, behavioural analysis, network analysis

Degrees

Bachelor of Science in Computing Science