Machine Learning on Blockchain Data: A Systematic Mapping Study (2403.17081v1)

Abstract: Context: Blockchain technology has drawn growing attention in the literature and in practice. Blockchain technology generates considerable amounts of data and has thus been a topic of interest for Machine Learning (ML). Objective: The objective of this paper is to provide a comprehensive review of the state of the art on machine learning applied to blockchain data. This work aims to systematically identify, analyze, and classify the literature on ML applied to blockchain data. This will allow us to discover the fields where more effort should be placed in future research. Method: A systematic mapping study has been conducted to identify the relevant literature. Ultimately, 159 articles were selected and classified according to various dimensions, specifically, the domain use case, the blockchain, the data, and the machine learning models. Results: The majority of the papers (49.7%) fall within the Anomaly use case. Bitcoin (47.2%) was the blockchain that drew the most attention. A dataset consisting of more than 1.000.000 data points was used by 31.4% of the papers. And Classification (46.5%) was the ML task most applied to blockchain data. Conclusion: The results confirm that ML applied to blockchain data is a relevant and a growing topic of interest both in the literature and in practice. Nevertheless, some open challenges and gaps remain, which can lead to future research directions. Specifically, we identify novel machine learning algorithms, the lack of a standardization framework, blockchain scalability issues and cross-chain interactions as areas worth exploring in the future.

• The paper systematically maps 159 studies to uncover key ML applications on blockchain data.
• It emphasizes the use of ML models for anomaly detection, cryptocurrency price prediction, and smart contract vulnerability detection, particularly on Ethereum.
• The study identifies challenges such as data complexity and the absence of standard benchmarks, paving the way for future research.

Machine Learning on Blockchain Data: A Comprehensive Investigation

Introduction

Blockchain technology has witnessed exponential growth and adoption across various sectors. This growth is accompanied by a substantial increase in the volume of data blockchain networks generate, such as transaction histories and smart contract deployments. Given the rich insights that can be derived from blockchain data, there has been a growing interest in applying ML techniques to analyze this data, aiming to uncover patterns, predict market trends or detect anomalous behavior.

This blog post explores a systematic mapping paper that scrutinizes the application of machine learning on blockchain data. The paper classifies 159 selected research articles published between 2008 and 2023, identifying core use cases such as anomaly detection, cryptocurrency price prediction, smart contract vulnerability detection, performance prediction, and address classification.

Key Insights from the Study

Anomaly Detection

• Predominant Focus: The anomaly detection domain, comprising 49.7% of the paper's dataset, is a primary focus area for applying ML on blockchain data. This domain includes Ponzi scheme detection and phishing detection, primarily on the Ethereum platform.
• Advanced Techniques: Techniques employed for anomaly detection range from classic machine learning models, like Random Forest and XGBoost, to more advanced deep learning techniques and graph-based models.
• Ethereum's Prominence: Ethereum, due to its smart contract functionality, emerges as a notable blockchain for anomaly detection research. This suggests a heightened interest in safeguarding decentralized applications against fraud and scams.

Cryptocurrency Price Prediction

• Broad Data Analysis: Studies in this domain leverage a variety of data points including technological indicators, economic factors, and social media sentiment to predict cryptocurrency prices.
• Bitcoin and Ethereum: A majority of the research focuses on predicting the prices of Bitcoin and Ethereum, highlighting their significance in the cryptocurrency market.
• Combination of Models: Several studies utilize a combination of machine learning models to improve prediction accuracy, indicating the complexity of price prediction in the volatile cryptocurrency market.

Smart Contract Vulnerability Detection

• Ethereum's Smart Contracts: All studies related to smart contract vulnerability detection concentrate on Ethereum. This underscores the critical role of ensuring the security and reliability of smart contracts.
• Novel Approaches: Researchers propose innovative models like Eth2Vec and Bytecode matching, demonstrating the ongoing efforts to enhance the detection of vulnerabilities in smart contracts.

Performance Prediction

• Emerging Area: Performance prediction, including transaction throughput and gas price forecasting, represents an emerging area of interest. This reflects a growing focus on optimizing blockchain efficiency and user experience.

Address Classification

• De-Anonymization Efforts: Address classification studies, with a significant focus on Bitcoin, aim at de-anonymizing blockchain transactions. This highlights the tension between blockchain's pseudonymity feature and the need for transparency and security.

Future Directions and Challenges

The paper identifies several challenges and future research directions:

• Data Volume and Complexity: As blockchains evolve, the volume and complexity of blockchain data continue to surge, posing significant challenges in data processing and analysis.
• Standardization and Benchmarking: The lack of standardized frameworks and benchmarks for evaluating ML applications in blockchain poses a major hurdle. There's a pressing need for community-wide efforts to establish such benchmarks.
• Emerging Technologies and Interactions: The integration of blockchain with other technologies, like IoT and AI, opens new avenues for research. However, it also necessitates novel ML approaches tailored to these complex interactions.

Conclusion

The systematic mapping paper sheds light on the current state and potential of applying machine learning to blockchain data. With the blockchain landscape rapidly evolving, there's a pressing need for novel machine learning techniques that can effectively analyze and derive valuable insights from blockchain data. Moreover, fostering a culture of data sharing and standardization will be crucial for advancing research in this domain and addressing the challenges inherent to blockchain's ever-expanding data ecosystem.