Large Language Models for Cryptocurrency Transaction Analysis: A Bitcoin Case Study

2025-01-30

AI Summary

Existing cryptocurrency transaction graph analysis methods are computationally complex and ineffective at uncovering user behavioral patterns. Method: This paper introduces a novel large language model (LLM)-based paradigm for Bitcoin transaction graph analysis, featuring a three-tier evaluation framework—(i) foundational metric computation, (ii) behavioral feature summarization, and (iii) contextual behavioral interpretation. We propose LLM4TG, the first human-readable graph representation format tailored for LLMs, and CETraS, a connectivity-enhanced transaction graph sampling algorithm, integrating graph structural modeling, prompt engineering, and adaptive subgraph sampling. Contribution/Results: Experiments demonstrate high accuracy in few-shot foundational metric prediction and generation of trustworthy behavioral explanations; our method significantly outperforms baselines on behavioral summarization. This work provides the first systematic empirical validation of LLMs for on-chain behavioral understanding and establishes a new methodology for interpretable blockchain analytics.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Large Multimodal Models (LMMs)
• Game Theory and Economic Paradigms: Behavioral Game Theory

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Cryptocurrencies are widely used, yet current methods for analyzing transactions heavily rely on opaque, black-box models. These lack interpretability and adaptability, failing to effectively capture behavioral patterns. Many researchers, including us, believe that Large Language Models (LLMs) could bridge this gap due to their robust reasoning abilities for complex tasks. In this paper, we test this hypothesis by applying LLMs to real-world cryptocurrency transaction graphs, specifically within the Bitcoin network. We introduce a three-tiered framework to assess LLM capabilities: foundational metrics, characteristic overview, and contextual interpretation. This includes a new, human-readable graph representation format, LLM4TG, and a connectivity-enhanced sampling algorithm, CETraS, which simplifies larger transaction graphs. Experimental results show that LLMs excel at foundational metrics and offer detailed characteristic overviews. Their effectiveness in contextual interpretation suggests they can provide useful explanations of transaction behaviors, even with limited labeled data.

Problem

• Cryptocurrency Transaction Analysis
• Pattern Recognition
• Behavior Analysis

Innovation

• Large Language Models
• CETraS Algorithm
• LLM4TG Graph Representation