Gyuyeon Na, Minjung Park, Hyeonjeong Cha, Soyoun Kim · 9 authors
Blockchain transaction networks are complex, with evolving temporal patterns and inter-node relationships. To detect illicit activities, we propose a hybrid GCN-GRU model that captures both structural and sequential features. Using real Bitcoin transaction data (2020-2024), our model achieved 0.9470 Accuracy and 0.9807 AUC-ROC, outperforming all baselines.
With the shift from Centralized Finance (CeFi) to Decentralized Finance (DeFi), financial transactions have become trustless and self-executing through blockchain platforms, creating new opportunities while exposing the ecosystem to significant fraud risks. However, due to the lack of centralized oversight and the vulnerabilities in the blockchain platforms, DeFi transactions still face several security challenges, including fraud, identity theft, insider threats, and data breaches. Various methods, including regulatory frameworks, machine learning (ML), and deep learning (DL) techniques, are employed to detect these threats, particularly fraud, in DeFi transactions. Although these approaches help identify fraudulent activities, they face challenges related to accuracy and zero-day attacks due to insufficient data and the complexity of emergingfraud patterns. This study presents a novel approach for detecting and profiling fraud attacks, including zero-day ones in DeFi transactions, thereby eliminating the reliance on wallet transaction history, a limitation that previous research has heavily depended on. The proposed approach leverages two key components: a novel analyzer named DeFiTransLyzer (V1.0) and an Advanced Genetic Algorithm (AGA) for fraud transaction profiling. DeFiTransLyzer extracts 79 features from transaction and wallet data. At the same time, the AGA incorporates advanced techniques, including Penalized Fitness Evaluation, Elite Retention Strategy, Dynamic Mutation Rate, and dynamic generation, to create precise fraud profiles. By focusing solely on transaction features, the model ensures that all fraudulent activities, including zero-day ones, initiated within the first transaction of a new account can be effectively detected, without relying on prior wallet activity. To address the scarcity of comprehensive validation datasets, we introduce BCCCDeFiFraudTrans-2025, which comprises 1,026,867 annotated Ethereum transaction samples from the DeFi ecosystem. Additionally, the study establishes two taxonomies for systematic classification, covering the literature on fraud detection and profiling methods. Experimental results demonstrate that the proposed method achieves superior accuracy, precision, and efficiency while offering interpretability through its profiling mechanism. These promising outcomes highlight the potential of AGA profiling to enhance the detection and identification of fraudulent activities, including zero-day ones within DeFi transactions, contributing to the security and resilience of blockchainbased financial systems.
As Ethereum continues to gain traction as a leading blockchain platform, its open and decentralized nature has also made it an attractive target for fraudulent activities. This paper presents a machine learning-based approach to detect fraudulent Ethereum transactions by analyzing behavioral patterns within transaction data. Using a labeled dataset of Ethereum transactions, various classification algorithms such as Random Forest, XGBoost, and Support Vector Machines were trained and evaluated. The proposed system focuses on identifying anomalies and suspicious transaction behavior by extracting relevant features like gas usage, transaction value, and timing. Experimental results show that the model can achieve high accuracy and precision in distinguishing between legitimate and fraudulent transactions. This work contributes to the growing field of blockchain security by demonstrating the viability of intelligent fraud detection techniques and providing a framework that can be integrated into real-world applications.
Smart contracts play a pivotal role in blockchain ecosystems, and fuzzing remains a critical approach to securing them. However, existing smart contract fuzzers often optimize either seed generation or mutation scheduling in isolation and rely on narrow, fragmented feedback signals, leaving multi-transaction reasoning and stagnation recovery under-explored. In this work, we propose aLarge Language Models(LLMs)-based Multi-feedback Smart Contract Fuzzing framework (LLAMA). Key components of the proposed LLAMA include: (i) a hierarchical prompting strategy that guides LLMs to generate structurally valid, context-aware multi-transaction initial seeds, together with a lightweight pre-fuzzing phase that validates and prioritizes high-potential LLM-generated candidates; (ii) a multi-feedback-guided evolutionary optimization module that jointly optimizes seed selection and mutation scheduling by a group of constraints for driving an LLM-bootstrapped bandit scheduler. (iii) an LLM-guided hybrid fuzzing module that integrates evolutionary fuzzing with a dual-channel recovery mechanism, which concurrently employs asynchronous coverage-stagnation- based LLM reseeding and selective symbolic execution to resolve complex path constraints. Our extensive experiments demonstrate that LLAMA outperforms state-of-the-art fuzzers in both coverage and vulnerability detection. Specifically, it achieves 92% instruction coverage on small contracts and 81% on large contracts, while detecting 132 out of 148 known vulnerabilities across diverse categories. Ablation studies further evidence that the proposed multi-feedback and hybrid recovery strategies have strong impact on LLAMA’s performance. The results explain LLAMA’s effectiveness, adaptability, and practicality in complex smart contract scenarios.
Survival modeling predicts the time until an event occurs and is widely used in risk analysis; for example, it's used in medicine to predict the survival of a patient based on censored data. There is a need for large-scale, realistic, and freely available datasets for benchmarking artificial intelligence (AI) survival models. In this paper, we derive a suite of 16 survival modeling tasks from publicly available transaction data generated by lending of cryptocurrencies in Decentralized Finance (DeFi). Each task was constructed using an automated pipeline based on choices of index and outcome events. For example, the model predicts the time from when a user borrows cryptocurrency coins (index event) until their first repayment (outcome event). We formulate a survival benchmark consisting of a suite of 16 survival-time prediction tasks (FinSurvival). We also automatically create 16 corresponding classification problems for each task by thresholding the survival time using the restricted mean survival time. With over 7.5 million records, FinSurvival provides a suite of realistic financial modeling tasks that will spur future AI survival modeling research. Our evaluation indicated that these are challenging tasks that are not well addressed by existing methods. FinSurvival enables the evaluation of AI survival models applicable to traditional finance, industry, medicine, and commerce, which is currently hindered by the lack of large public datasets. Our benchmark demonstrates how AI models could assess opportunities and risks in DeFi. In the future, the FinSurvival benchmark pipeline can be used to create new benchmarks by incorporating more DeFi transactions and protocols as the use of cryptocurrency grows.
Hong Qu, Krzysztof Gogol, Florian Grötschla, Claudio J. Tessone
Decentralized Finance (DeFi) lending enables permissionless borrowing via smart contracts. However, it faces challenges in optimizing interest rates, mitigating bad debt, and improving capital efficiency. Rule-based interest-rate models struggle to adapt to dynamic market conditions, leading to inefficiencies. This work applies Offline Reinforcement Learning (RL) to optimize interest rate adjustments in DeFi lending protocols. Using historical data from Aave protocol, we evaluate three RL approaches: Conservative Q-Learning (CQL), Behavior Cloning (BC), and TD3 with Behavior Cloning (TD3-BC). TD3-BC demonstrates superior performance in balancing utilization, capital stability, and risk, outperforming existing models. It adapts effectively to historical stress events like the May 2021 crash and the March 2023 USDC depeg, showcasing potential for automated, real-time governance.
Abstract— This study integrates blockchain technology and machine learning to enhance credit card fraud detection. Precise fraud prediction is performed using advanced algorithms such as Random Forest, Logistic Regression, XGBoost, and Bayesian models. Tools such as Ganache and MetaMask from Ethereum blockchain facilitate safe and transparent tracking of suspicious transactions. Decentralized and tamper-proof properties of blockchain add reliability, and machine learning adds precision and flexibility. The system is highly accurate and transparent and has the potential to be used to fight financial fraud. Keywords— Credit Card Fraud, Blockchain, Machine Learning, Ethereum, Web3, SMOTE, XGBoost, Streamlit
The rise of digital finance has led to a surge in fraudulent activities, particularly in credit card transactions and cryptocurrency ecosystems. With financial crimes becoming more sophisticated, traditional fraud detection methods often fail to identify complex fraudulent patterns. This research explores the application of machine learning (ML) and artificial intelligence (AI) techniques to enhance the security of digital finance by detecting fraudulent activities in credit card transactions and cryptocurrency wallets within the USA. The study utilizes large-scale transaction datasets containing key financial indicators such as transaction frequency, spending patterns, anomaly scores, and network behaviors. To develop an AI-driven fraud detection framework, we implement and compare six machine learning models: XGBoost, RLightGBM, Decision Trees, K-Nearest Neighbors (KNN), Convolutional Neural Networks (CNNs), and Autoencoders. The models are trained on both structured financial data (e.g., credit card transaction logs) and unstructured blockchain transaction records (e.g., Bitcoin wallet addresses and transaction flows). To address data imbalance, the study applies the Synthetic Minority Over-sampling Technique (SMOTE), ensuring fair representation of fraudulent transactions. Model performance is evaluated using Precision, Recall, F1-score, and ROC-AUC metrics to determine the most effective fraud detection approach. Additionally, the research emphasizes data privacy and security, incorporating anonymization techniques and regulatory compliance measures to safeguard sensitive financial information. This study contributes to the ongoing fight against financial fraud by demonstrating how AI-based solutions can enhance the security and resilience of digital finance systems in the USA.
Dan Lin, Ziye Zheng, Jiajing Wu, Jingjing Yang · 8 authors
Cross-chain technology enables seamless asset transfer and message-passing within decentralized finance (DeFi) ecosystems, facilitating multi-chain coexistence in the current blockchain environment. However, this development also raises security concerns, as malicious actors exploit cross-chain asset flows to conceal the provenance and destination of assets, thereby facilitating illegal activities such as money laundering. Consequently, the need for cross-chain transaction traceability has become increasingly urgent. Prior research on transaction traceability has predominantly focused on single-chain and centralized finance (CeFi) cross-chain scenarios, overlooking DeFispecific considerations. This paper proposes ABCTRACER, an automated, bi-directional cross-chain transaction tracing tool, specifically designed for DeFi ecosystems. By harnessing transaction event log mining and named entity recognition techniques, ABCTRACER automatically extracts explicit cross-chain cues. These cues are then combined with information retrieval techniques to encode implicit cues. ABCTRACER facilitates the autonomous learning of latent associated information and achieves bidirectional, generalized cross-chain transaction tracing. Our experiments on 12 mainstream cross-chain bridges demonstrate that ABCTRACER attains 91.75% bi-directional traceability (F1 metrics) with self-adaptive capability. Furthermore, we apply ABCTRACER to real-world cross-chain attack transactions and money laundering traceability, thereby bolstering the traceability and blockchain ecological security of DeFi bridging applications.
By combining blockchain technology, machine learning, and artificial intelligence (AI), the banking sector has witnessed a revolution in credit risk reduction in recent years. With an emphasis on predictive analytics and decentralized frameworks, this paper explores the real-world applications of these technologies in the discovery, evaluation, and management of credit risk. The study demonstrates how machine learning models, blockchain's transparent and unchangeable ledger systems, and AI-powered algorithms have greatly increased the precision and effectiveness of credit risk assessments through thorough literature analysis and case studies. The report also examines how financial institutions implement these technologies to improve operational risk management, lower fraud, and create more accurate credit scoring systems. Notwithstanding their promise, there are still significant obstacles to overcome, including data privacy, regulatory compliance, and implementation costs. In order to effectively utilize the advantages of AI, blockchain, and machine learning in reducing credit risk, the article ends with ideas for overcoming these obstacles. Keywords: Artificial Intelligence; Blockchain; Machine Learning; Credit Risk Mitigation; Predictive Analytics; Financial Technology; Credit Scoring; Risk Management; Decentralized Finance; Operational Risk
Decentralized Finance (DeFi) lending protocols currently rely on fixed collateralization ratios, leading to inefficiencies such as over-collateralization, frequent liquidations, and suboptimal capital utilization. This paper proposes a novel framework integrating machine learning (ML) with DeFi lending protocols to dynamically adjust collateral requirements in realtime based on borrower behavior, market volatility, and on-chain data. By analyzing historical loan performance, social sentiment, and macroeconomic indicators, the ML model optimizes collateral ratios to minimize liquidations while maintaining protocol security. We simulate the model using data from major DeFi platforms (e.g., Aave, Compound) and demonstrate a 30-50
Muhammad Shoaib Farooq, Shahzada Fahad Munir, Muhammad Faraz Manzoor, Momina Shaheen
The increasing complexity of fraudulent activities requires advanced fraud detection systems, as existing solutions lack effectiveness due to two challenges. First, privacy concerns prevent financial institutions from sharing sensitive transaction data. Second, data imbalance causes biased models, as fraudulent transactions represent a small fraction of total transactions, leading to poor fraud detection performance. To address these challenges, we propose an AI‐driven adaptive federated learning (AFL) framework for credit card fraud detection (CCFD). AFL enables decentralized learning, allowing financial institutions to train a global fraud detection model collaboratively without sharing raw transaction data. The model aggregation is performance‐adaptive, weighting client contributions based on detection accuracy to ensure a robust global model. To overcome data imbalance, we introduce a multistep data balancing framework integrating Tomek links for undersampling, borderline‐SMOTE for oversampling, and cognitive sample pruning to remove misleading samples. To evaluate the robustness and generalizability of the proposed framework, we conducted experiments on both the widely used 2013 Kaggle dataset and the Sparkov simulated dataset (2019‐2020). The Sparkov dataset, which contains interpretable demographic and merchant‐level features, allowed us to test the model’s adaptability to diverse data sources. The results demonstrate that the proposed AFL framework, combined with advanced data balancing, significantly outperforms traditional models, achieving 99% accuracy, 99.5% precision, 99.4% recall, and 99% F1‐score on the Kaggle dataset, and 97.4% accuracy, 99.5% precision, 97.5% recall, and 98.4% F1‐score on the Sparkov dataset. This research highlights AI’s transformative role in finance, particularly in enhancing fraud detection systems with improved accuracy, robustness, security, and scalability.
Naresh Kumar Satish, Mathieu Mercadier, Cristina Hava Muntean, Anderson Augusto Simiscuka
The cryptocurrency market is widely regarded as one of the most volatile financial markets due to inconsistencies in its pricing factors. Despite this volatility, it continues to attract a large population of investors, many of whom incur significant losses. To address this challenge and support risk assessment for investors, users, and other stakeholders, this paper focuses on forecasting Ethereum prices by analyzing social media sentiment. The study gathers data from sources such as global news headlines and Reddit discussion forums, enhancing it with hybrid sentiment features derived from the VADER, BERT and TextBlob models. These sentiment insights are then correlated with Ethereums financial parameters to establish meaningful relationships within the data, which are used to train machine learning models. The study evaluates the predictive performance of Random Forest, Extreme Gradient Boosting, and Long Short-Term Memory models. Among these, Extreme Gradient Boosting demonstrated superior performance, effectively capturing complex relationships within the data and achieving an R-squared value of 0.982115. To further enhance the studys risk assessment capabilities, the concept of Explainable Artificial Intelligence (XAI) is employed to improve transparency and accountability in the model outcomes. Specifically, Shapley Additive Explanations (SHAP) are used to interpret the feature interactions within the Extreme Gradient Boosting model, thereby increasing its reliability and providing deeper insights into its decision-making process.
Illicit transaction detection on blockchain networks presents a critical challenge due to the pseudonymous, decentralized, and high-volume nature of decentralized finance (DeFi) ecosystems. Traditional machine learning models struggle to effectively capture the temporal dynamics and irregular patterns of illicit behavior, while graph-based methods often incur high computational costs and rely on static relational structures. In this paper, we propose a novel dual-attention framework—GAM-CoT Transformer—for robust transaction-level anomaly detection. The proposed model integrates two key components: a Global Attention Module (GAM) that adaptively reweights feature channels and temporal steps to emphasize salient patterns, and a Contextual Transformer (CoT) block that efficiently models short-range dependencies using grouped convolutions instead of full self-attention. This design enables the model to simultaneously achieve computational efficiency, temporal expressiveness, and improved detection sensitivity. We evaluate our approach on a real-world blockchain transaction dataset and demonstrate its superiority over conventional classifiers including Random Forest, XGBoost, and LSTM-based models. The GAM-CoT Transformer achieves higher recall and F1 scores, particularly for the minority illicit class, while maintaining fast convergence and deployment scalability. Our method offers a practical and effective solution for enhancing the security of blockchain systems through intelligent transaction behavior modeling.
Smart contracts are self-executing programs on blockchains that manage complex business logic with transparency and integrity. However, their immutability after deployment makes programming errors particularly critical, as su... | Find, read and cite all the research you need on Tech Science Press
As decentralized finance (DeFi) expands, Ethereum’s role as the backbone for digital asset exchange, smart contracts, and financial protocols has grown—but so has its exposure to fraud. Phishing, money laundering, and malicious contracts exploit its openness. Existing ML and deep learning models often lack the balance between speed, accuracy, and explainability needed for real-time blockchain analysis. High-performing models like transformers are accurate but too resource-heavy and opaque. This study leverages TabNet—a sparse-attention deep learning model optimized for tabular Ethereum transaction data. It dynamically selects relevant features during training, enhancing both interpretability and efficiency. With an accuracy of 0.86, precision of 0.80, and F1-score of 0.79, TabNet outperforms traditional models in fraud detection while remaining lightweight and transparent. Its feature-level insights make it ideal for environments where trust, latency, and transparency are crucial. The results position TabNet as a scalable, practical alternative for fraud detection in blockchain ecosystems.
Effective risk management solutions become absolutely crucial when financial markets embrace distributed technology and decentralized financing (DeFi). This study offers a thorough survey and comparative analysis of the integration of artificial intelligence (AI) in risk management for distributed arbitrage systems. We examine several modern caching techniques namely in memory caching, distributed caching, and proxy caching and their functions in enhancing performance in decentralized settings. Through literature review we examine the utilization of AI techniques for alleviating risks related to market volatility, liquidity challenges, operational failures, regulatory compliance, and security threats. This comparison research evaluates various case studies from prominent DeFi technologies, emphasizing critical performance metrics like latency reduction, load balancing, and system resilience. Additionally, we examine the problems and trade offs associated with these technologies, emphasizing their effects on consistency, scalability, and fault tolerance. By meticulously analyzing real world applications, specifically centering on the Aave platform as our principal case study, we illustrate how the purposeful amalgamation of AI with contemporary caching methodologies has revolutionized risk management in distributed arbitrage systems.
This study presents a comprehensive systematic review of Artificial Intelligence (AI) applications in DecentralizedFinance (DeFi), emphasizing AI’s pivotal role in mitigating the vulnerabilities and operational complexities inherentin permissionless financial systems. By systematically analyzing 39 peer-reviewed studies from major scholarlydatabases, the review identifies five dominant application domains: fraud detection, smart contract security, marketprediction, credit risk assessment, and decentralized governance. It examines the diverse range of AI methodsspanning machine learning, deep learning, graph neural networks, and reinforcement learning—and evaluates theircomparative performance and limitations. The findings reveal that AI not only enhances DeFi’s transparency, trust,and efficiency but also underpins emerging capabilities such as autonomous governance and adaptive marketmechanisms. Persistent challenges including data scarcity, cross-chain generalization, interpretability, andscalability—underscore the need for robust, explainable, and ethical AI solutions. The review concludes that AIconstitutes a foundational enabler for secure, transparent, and resilient decentralized financial ecosystems, andoutlines critical future research directions for integrating trustworthy intelligence into the evolving DeFi landscape. (PDF) A Systematic Review on the Application of Artificial Intelligence in Decentralized Finance. Available from: https://www.researchgate.net/publication/397514996_A_Systematic_Review_on_the_Application_of_Artificial_Intelligence_in_Decentralized_Finance [accessed Nov 11 2025].
This research explores the factors contributing to the failure of cryptocurrency exchanges by analyzing a sample of 845 exchanges. Using logit and probit models, it identifies key variables affecting cryptocurrency exchange defaults. The results show that cryptocurrency exchanges that are centralized, located in countries with high transparency indices, and offer fewer peer cryptocurrencies are more likely to default. Additionally, exchanges that impose high withdrawal fees and have no restrictions on clients from the United States are also positively associated with defaults. Moreover, the absence of referral schemes and having lower ratings each contributes marginally to defaults. Machine learning (ML) models including random forest, support vector machine, stacked ensemble confirm the robustness and high predictability of cryptocurrency exchange defaults. • This study uses statistical and ML models to predict cryptocurrency exchange default. • Centralized exchanges from high-transparency-index nations are more prone to default. • Limited coin listings, high fees, and U.S. client access increase the default risk. • Lacking referral programs and low ratings both marginally contribute to default.