Blockchain Intelligence: Methods, Applications and Challenges

✍ Scribed by Zibin Zheng (editor), Hong-Ning Dai (editor), Jiajing Wu (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 170
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

This book focuses on using artificial intelligence (AI) to improve blockchain ecosystems. Gathering the latest advances resulting from AI in blockchain data analytics, it also presents big data research on blockchain systems.

Despite blockchain's merits of decentralisation, immutability, non-repudiation and traceability, the development of blockchain technology has faced a number of challenges, such as the difficulty of data analytics on encrypted blockchain data, poor scalability, software vulnerabilities, and the scarcity of appropriate incentive mechanisms. Combining AI with blockchain has the potential to overcome the limitations, and machine learning-based approaches may help to analyse blockchain data and to identify misbehaviours in blockchain. In addition, deep reinforcement learning methods can be used to improve the reliability of blockchain systems.

This book focuses in the use of AI to improve blockchain systems and promote blockchain intelligence. It describes data extraction, exploration and analytics on representative blockchain systems such as Bitcoin and Ethereum. It also includes data analytics on smart contracts, misbehaviour detection on blockchain data, and market analysis of blockchain-based cryptocurrencies. As such, this book provides researchers and practitioners alike with valuable insights into big data analysis of blockchain data, AI-enabled blockchain systems, and applications driven by blockchain intelligence.

✦ Table of Contents

Preface
Acknowledgments
Contents
About the Editors
1 Overview of Blockchain Intelligence
1.1 Overview
1.2 Blockchain
1.3 Smart Contract
1.4 Blockchain Intelligence
1.4.1 Limitations of Blockchain and Smart Contracts
1.4.2 Opportunities Brought by Blockchain Intelligence
1.5 Summary
References
2 On-chain and Off-chain Blockchain Data Collection
2.1 Overview
2.2 Ethereum and Smart Contracts
2.2.1 Peer and Blockchain
2.2.2 Smart Contract
2.2.3 Tokens and Clients
2.3 Raw Data Extraction from Ethereum
2.3.1 Block
2.3.2 Trace
2.3.3 Receipt
2.4 Data Exploration of Ethereum
2.4.1 Dataset 1: Block and Transaction
2.4.2 Dataset 2: Internal Ether Transaction
2.4.3 Dataset 3: Contract Info
2.4.4 Dataset 4: Contract Call
2.4.5 Dataset 5: ERC20 Token Transaction
2.4.6 Dataset 6: ERC721 Token Transaction
2.5 Applications of XBlock-ETH
2.5.1 Blockchain System Analysis
2.5.1.1 Decentralization Analysis
2.5.1.2 Gasprice Prediction
2.5.1.3 Performance Benchmark
2.5.2 Smart Contract Analysis
2.5.2.1 Contract Similarity and Recommendation
2.5.2.2 Contract Developer Analysis
2.5.2.3 Contract Vulnerability Detection
2.5.2.4 Fraud Detection
2.5.3 Cryptocurrency Analysis
2.5.3.1 Cryptocurrency Transferring Analysis
2.5.3.2 Cryptocurrency Price Analysis
2.5.3.3 Fake User Detection
2.6 Summary
References
3 Analysis and Mining of Blockchain Transaction Network
3.1 Overview
3.2 Basic Knowledge of Network
3.2.1 Concept of Network
3.2.2 Mathematical Representation of Network
3.3 Blockchain Transaction Network
3.3.1 Method of Modeling Blockchain Transaction Network
3.3.2 Modeling Ethereum Transaction Network by Graph
3.3.2.1 Data Collection
3.3.2.2 Network Construction
3.4 Data Analysis and Mining Based on Blockchain Transaction Network
3.4.1 Temporal Weighted Multidigraph Embedding
3.4.1.1 Random Walk
3.4.1.2 Learning Process
3.4.2 Phishing Scam
3.4.2.1 Data Acquisition
3.4.2.2 Setting
3.4.2.3 Metrics
3.4.2.4 Results
3.4.3 Link Prediction
3.4.3.1 Data Acquisition
3.4.3.2 Setting
3.4.3.3 Results
3.4.4 Transaction Tracking on Blockchain
3.4.4.1 Embedding Based Link Prediction for Investigation
3.4.4.2 Evaluation Measurement of Temporal Link Prediction
3.4.5 Results and Analysis
3.4.5.1 Investigation on the Transaction Time
3.4.5.2 Investigation on the Transaction Amount
3.4.5.3 Investigation on Both Factors
3.5 Summary
References
4 Intelligence-Driven Optimization of Smart Contracts
4.1 Overview
4.2 Smart Contracts Similarity Analysis
4.2.1 Syntax Similarity Analysis
4.2.2 Semantic Similarity Analysis
4.2.3 Similarity Calculation of Smart Contracts
4.3 Differentiated Code Recommendation
4.3.1 Similar Smart Contracts Clustering
4.3.2 Differentiated Code Extraction
4.4 Case Study
4.4.1 Dataset
4.4.2 Research Questions
4.4.3 Evaluation Criteria
4.4.4 Results Analysis
4.4.4.1 RQ1
4.4.4.2 RQ2
4.4.4.3 RQ3
4.4.5 Cluster Analysis
4.5 Discussion
4.5.1 Related Work
References
5 Misbehavior Detection on Blockchain Data
5.1 Overview
5.2 Data Analysis on Blockchain
5.2.1 Analysis Based on Complex Network
5.2.2 Analysis Based on Data Mining Methods
5.2.3 Analysis Based on Statistic Tools
5.3 Case Study: Ponzi Scheme Detection
5.3.1 Ponzi Scheme on Ethereum
5.3.2 Ethereum and Smart Contracts
5.3.2.1 A Source Code Snippet of a Smart Ponzi Scheme
5.3.2.2 Deploy a Contract
5.3.3 Data, Feature Extraction, and Classification Model
5.3.3.1 Data
5.3.3.2 Account Features
5.3.3.3 Code Features
5.3.3.4 Classification Model
5.3.4 Experimental Results and Feature Analysis
5.3.4.1 Experiment Setting
5.3.4.2 Results Summary
5.3.4.3 Feature Analysis
5.3.4.4 Application
5.3.5 Related Work
5.3.6 Future Work
5.4 Case Study: Phishing Scam Detection
5.4.1 Phishing Scams on Ethereum
5.4.2 Background and Related Work
5.4.3 Proposed Method
5.4.3.1 Cascade Feature Extraction Method
5.4.3.2 Dual-Sampling Ensemble Method
5.4.4 Data Collection and Preparation
5.4.4.1 Data Collection
5.4.4.2 Data Cleaning
5.4.5 Experiment Result and Analysis
5.4.5.1 Experiment Settings
5.4.5.2 Method Comparison
5.4.5.3 Example Sampling Effect Analysis
5.4.5.4 Feature Sampling Evaluation
5.4.5.5 Feature Analysis
5.4.6 Conclusion and Future Work
5.5 Summary
References
6 Market Analysis of Blockchain-Based Cryptocurrencies
6.1 Overview
6.2 Features Analysis on Cryptocurrencies Market
6.2.1 Introduction
6.2.2 Related Work
6.2.3 Methods
6.2.3.1 Detrended Fluctuation Analysis
6.2.3.2 A-MFDFA Method
6.2.3.3 Causality-in-Quantiles Test
6.2.4 Data and Empirical Findings
6.2.4.1 Data
6.2.5 Results
6.2.5.1 Long-Range Dependence
6.2.5.2 Multi-fractality
6.2.5.3 Causality
6.3 Detecting Abnormal Schemes on Cryptocurrencies Market
6.3.1 Introduction
6.3.2 Dataset and the Algorithm
6.3.2.1 Dataset
6.3.2.2 Improved Apriori Algorithm
6.4 Experimental Results
6.4.1 Abnormal Trading Behavior
6.4.2 Abnormal Trading Price
6.5 Summary
References
7 Open Research Directions
7.1 Overview
7.2 Federated Learning on Blockchain
7.3 Collective Intelligence Bestowing Blockchain
7.3.1 Collective Intelligence Bestowing Blockchain Systems
7.3.2 Collective Intelligence Bestowing Smart Contracts
7.4 Artificial Intelligence to Enhance Blockchain Automation
7.4.1 Intelligent Operational Management of Blockchain Systems
7.4.2 AI-Empowered Scalable Blockchain Systems
References