With the increasing popularity of Ethereum, smart contracts have become a prime target for fraudulent activities such as Ponzi, honeypot, gambling, and phishing schemes. While some researchers have studied intelligent fraud detection, most research has focused on identifying Ponzi contracts, with little attention given to detecting and preventing gambling or phishing contracts. There are three main issues with current research. Firstly, there exists a severe data imbalance between fraudulent and non-fraudulent contracts. Secondly, the existing detection methods rely on diverse raw features that may not generalize well in identifying various classes of fraudulent contracts. Lastly, most prior studies have used contract source code as raw features, but many smart contracts only exist in bytecode. To address these issues, we propose a fraud detection method that utilizes Efficient Channel Attention EfficientNet (ECA-EfficientNet) and data enhancement. Our method begins by converting bytecode into Red Green Blue (RGB) three-channel images and then applying channel exchange data enhancement. We then use the enhanced ECA-EfficientNet approach to classify fraudulent smart contract RGB images. Our proposed method achieves high F1-score and Recall on both publicly available Ponzi datasets and self-built multi-classification datasets that include Ponzi, honeypot, gambling, and phishing smart contracts. The results of the experiments demonstrate that our model outperforms current methods and their variants in Ponzi contract detection. Our research addresses a significant problem in smart contract security and offers an effective and efficient solution for detecting fraudulent contracts.
Blockchain technology has piqued the interest of businesses of all types, while consistently improving and adapting to business requirements. Several blockchain platforms have emerged, making it challenging to select a suitable one for a specific type of business. This paper presents a classification of over one hundred blockchain platforms. We develop smart contracts for detecting healthcare insurance frauds using the top two blockchain platforms selected based on our proposed decision-making map approach which selects the top suitable platforms for healthcare insurance frauds detection application. Our classification shows that the largest percentage of platforms can be used for all types of application domains, the second biggest percentage for financial services, and a small number is to develop applications in specific domains. Our decision-making map and performance evaluations reveal that Hyperledger Fabric surpassed Neo in all metrics for detecting healthcare insurance frauds.
Open access
Blockchain Technology Applications and Security
Imbalanced Data Classification Techniques
Artificial Intelligence in Healthcare and Education
Cryptocurrency has emerged as a decentralized transaction to overcome the problems of the centralized transaction system. Although it has become a popular trend in online cryptocurrency transactions and mobile wallets, this method has increased the number of fraudulent transactions instead of physically transferring money. Because the shared data and the history of online transactions may lead to fraudulent transactions. The preprocess identification of fraudulent cryptocurrency transactions is becoming an urgent research question. With the exponential blossoming of Artificial Intelligence, the employing of deep learning in predicting social issues has been achieved in many disciplines. From this perspective, this paper proposes an ensemble learning approach for fraudulent cryptocurrency transactions by integrating two deep learning methods: Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). The off-the-shelf CNN and LSTM, ensemble CNN, and ensemble LSTM with the bagged and boosted approach are compared in terms of accuracy and losses from training and test datasets. Moreover, the 10-fold cross-validation approach is employed for the evaluation of the proposed approach. The evaluation results indicate that the bagged LSTM ensembled approach is significant with 96.4% accuracy and outperforms the other approaches.
Bitcoin has a reputation of being used for unlawful activities, such as money laundering, dark web transactions, and payments for ransomware in the context of smart cities. Blockchain technology prevents illegal transactions, but cannot detect these transactions. Anomaly detection is a fundamental technique for recognizing potential fraud. The heuristic and signature-based approaches were the foundation of earlier detection techniques, but tragically, these methods were insufficient to explore the entire complexity of anomaly detection. Machine Learning (ML) is a promising approach to anomaly detection, as it can be trained on large datasets of known malware samples to identify patterns and features of the transactions. Researchers are focusing on determining an efficient fraud and security threat detection model that overcomes the drawbacks of the existing methods. Therefore, ensemble learning can be applied to anomaly detection in Bitcoin by combining multiple ML classifiers. In the proposed model, the ADASYN-TL (Adaptive Synthetic + Tomek Link) balancing technique is used for data balancing. Random search, grid search and Bayesian optimization are used for hyperparameter tuning. The hyperparameters have a great impact on the performance of the model. For classification, we used the stacking model by combining Decision Tree, Naive Bayes, K-Nearest Neighbors, and Random Forest. We used SHapley Additive exPlanation (SHAP) to interpret the predictions of the stacking model. The model also explores the performance of different classifiers using accuracy, F1-score, Area Under Curve-Receiver Operating Characteristic (AUC-ROC), precision, recall, False Positive Rate (FPR) and execution time, and ultimately selects the ideal model. The proposed model contributes to the development of effective fraud detection models that address the limitations of the existing algorithms. Our stacking model, which combines the prediction of multiple classifiers, achieved the highest F1-score of 97%, precision of 96%, recall of 98%, accuracy of 97%, AUC-ROC of 99% and FPR of 3%.
Due to the anonymity of blockchain, frequent security incidents and attacks occur through it, among which the Ponzi scheme smart contract is a classic type of fraud resulting in huge economic losses. Machine learning-based me... | Find, read and cite all the research you need on Tech Science Press
In recent years, with the great success of pre-trained language models, the pre-trained BERT model has been gradually applied to the field of source code understanding. However, the time cost of training a language model from... | Find, read and cite all the research you need on Tech Science Press
M. Mazhar Rathore, Sushil S. Chaurasia, Dhirendra Shukla
Anonymity is one of major factors that is causing the rise of bitcoin crypto-currency. There are several attacks (positive or negative) to de-anonymize the bitcoin addresses, in order to link the bitcoin entity to a physical entity or person. Bitcoin mixing service (called mixer) is one of the approaches to keep the user’s crpto-anonymity in the transparent ledger of bitcoin network. Mixers breaks the link between the sender and the receiver by mixing up coins received from multiple sources, while creating a mess to make it impossible to identify the actual sender of bitcoins. On the other hand, mixing services are being vastly exploited by criminals for laundering the illegal money, taken from frauds, ransom, scams, or other illegal activities. Detecting mixing services or mixer’s involvement in a transaction can help in discovering money laundering activities in the bitcoin blockchain. Existing mixer’s detection approaches either have a low accuracy-rate due to the changing nature of the mixing process or they are not efficient enough to be implemented in a real-time environment. In this paper, we developed a highly accurate decision-tree based model using C4.5 machine learning approach to identify addresses providing mixing services. To make this detection process efficient and be able to work in a real environment, we reduced overall feature-set to only eight features, minimizing overall computation time. Further, we shrink the decision-tree using reduced error-pruning to make the detection process faster. With the short decision-tree-size of 55 nodes, we achieved the accuracy of more than 97%, which is quite higher.
Kowshik Sankar Roy, Md. Ebtidaul Karim, Pritom Biswas Udas
The extensive usage of the Blockchain technology as one of the most popular forms of decentralized platform has been spread across a numerous field over the recent years. From financial sectors like banking industry to the supply chain management of multiple large corporate farms, blockchain technology has been proven its productivity across different communities. However, the reliability of the blockchain system has often been compromised with the introduction of various scams and fraudulent activity within the system. Due to the absence of a comprehensive and definitive dataset, the challenges of building an effective fraud detection model becomes even more sever in this particular field. Thus, in our work, we propose a deep learning based blockchain fraud detection model based on the Ethereum blockchain transaction data. With the association of a reliable dataset in this field, we build a deep learning-based detection model to classify the fraudulent activities within the system. The proposed classification model is comprised of a Long Short-Term Memory (LSTM) unit and a dense unit to detect the fraudulent transactions. For the sake of reducing down the complexity and avoiding the unnecessary transactional features, Information Gain has been utilized as the feature selection unit of the model. When compared to the corresponding values of different models on the same dataset, experimental results show a significant improved results in different aspects using the proposed approach.
The classification of the smart contract can effectively reduce the search space and improve retrieval efficiency. The existing classification methods are based on natural language processing technologies. Because the processing of source code by these technologies lacks extraction and processing in the software engineering field, there is still a lot of room for improvement in their methods of feature extraction. Therefore, this paper proposes a multi-feature fusion method for smart contract classification (MFF-SC) based on the code processing technology. From the source code perspective, source code processing method and attention mechanism are used to extract local code features. Structure-based traversal method are used to extract global code features from abstract syntax tree. Local and global code features introduce attention mechanism to generate code semantic features. From the perspective of account transaction, the feature of account transaction is extracted by using TransR. Next, the code semantic features and account transaction features generate smart contract semantic features by an attention mechanism. Finally, the smart contract semantic features are fed into a stacked denoising autoencoder and a softmax classifier for classification. Compared with other baselines and variants, experimental results on a real-world dataset show that MFF-SC achieves an accuracy rate of 83.9%.
With the rapid growth of Blockchain, Bitcoin, a key Blockchain application, has received a lot of attention. Bitcoin trading has made transactions more convenient. However, because of the anonymity, complexity, and lack of third parties, criminal activity against Blockchain Bitcoin applications is frequent. Individuals and society have suffered enormous losses as a result of money laundering, fraud, airdrop, ransom, and other peculiar abnormal transactions. We propose the GRU-GAT based model to detect abnormal transactions in Blockchain and reduce the loss and harm caused by abnormal transactions. Our proposed model employs a bidirectional recurrent neural network to extract the features of Blockchain Bitcoin transactions, graph attention networks for feature weighting, and the spatio-temporal aspects of transactions to integrate the features of Bitcoin transactions for anomaly identification. We conduct experiment through the publicly available elliptic dataset. The results reveal that the suggested method outperforms the comparison model in terms of accuracy and enhances the accuracy of detecting anomalous transactions in Blockchain digital currency.
Stefano Marzo, Royston Pinto, Lucy McKenna, Rob Brennan
Federated learning (FL) is a distributed machine learning<br> approach that enables remote devices i.e. workers to collaborate to compute<br> the fitting of a neural network model without sharing their data.<br> While this method is favorable to ensure data privacy, an imbalanced<br> data distribution can introduce unfairness in the model training, causing<br> discriminatory bias towards certain under-represented groups. In this paper,<br> we show that imbalance federated data decreases indexes of equity<br> i.e. differences in treatment for underrepresented classes. To address the<br> problem, we propose a federated learning framework called Z-Fed that 1)<br> balances the training without exchange of privacy protected data using<br> a zero knowledge proof (ZKP) technique, and 2) allows for the collection<br> of information on data distributions based on one or more categorical<br> features to produce metadata about population proportions. The proposed<br> framework infers the precise data distribution without exchanging<br> knowledge of the data categories and uses it to coordinate a balanced<br> training set. Z-Fed aims to mitigate the effect of imbalanced data in<br> FL while respecting privacy and without using mediators or probabilistic<br> approaches. Compared to a non-balanced framework, Z-Fed improves<br> fairness and equality measured in equal opportunities (EPD) by 53.54%,<br> equal odds (EOD) by 56.41%, and statistical parity (SPD) by 46.1% on<br> imbalanced UTK datasets, reducing biased predictions among subgroups.<br> EPD, EOD, and SPD measure the disparity of treatment between privileged<br> e.g. over-represented and non-privileged groups. Given the results<br> obtained, Z-Fed can reduce discriminatory behaviors and enhance trustworthy<br> of federated learning.
Open access
2 source records
Privacy-Preserving Technologies in Data
Imbalanced Data Classification Techniques
Artificial Intelligence in Healthcare and Education
Dec 1, 2022·2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys)
Bitcoin, due to its decentralization and anonymity, is more frequently utilized in cybercrimes such as money laundering, darknet market, and blackmail. In order to maintain the development of cryptocurrency communities, it is vital to fight against these cryptocurrency-related (e.g., Bitcoin) cybercrimes. To better monitor cryptocurrency-based cybercrime, it is essential for a deeper understanding of the patterns that exist in Bitcoin transactions. In this paper, we propose a novel framework consisting of empirical analysis and machine learning-based analysis for studying three typical categories of Bitcoin addresses from a dynamic perspective of subgraph evolution. Specifically, we collect 22,001 continuous blocks to build a Bitcoin transaction graph and extract subgraphs from the constructed graph for each address in different periods. In the empirical analysis, we have several important findings from three perspectives. While in the machine learning-based analysis, we propose a 110-dimensional feature set for classification tasks among these three types of Bitcoin addresses, where decision tree, k-nearest neighbor algorithm, and random forest are applied. The experimental results illustrate that metrics of all classifiers can achieve 0.88, and the highest accuracy rate reaches 0.95 in the random forest classifier. In addition, we explore the feature importance in the random forest using the SHAP library and discuss several interesting phenomena that appear in the evolution process of the subgraphs.
Lucas Penteado Lopes da Silva, Luiz Adeildo da Silva, Josafat Marinho Falcão Neto, Geidson Benício Coelho de Souza
Elaboração de algoritmos de aprendizado de máquina para a previsão do comportamento de preços da criptomoeda Ethereum, utilizando-se uma base de dados pública (Kaggle). Os modelos elaborados foram do tipo linear (ARIMA, séries temporais) e nãolinear (três modelos de redes neurais LTSM). Como melhor resultado, verificou-se que um dos modelos não lineares foi capaz de realizar previsões distantes em média de 4,32% dos preços reais.
Youwei Huang, Sen Fang, Jianwen Li, Bin Hu · 6 authors
In recent years, research in software security has concentrated on identifying vulnerabilities in smart contracts to prevent significant losses of crypto assets on blockchains. Despite early successes in this area, detecting developers' intents in smart contracts has become a more pressing issue, as malicious intents have caused substantial financial losses. Unfortunately, existing research lacks effective methods for detecting development intents in smart contracts. To address this gap, we propose \textsc{SmartIntentNN} (Smart Contract Intent Neural Network), a deep learning model designed to automatically detect development intents in smart contracts. \textsc{SmartIntentNN} leverages a pre-trained sentence encoder to generate contextual representations of smart contracts, employs a K-means clustering model to identify and highlight prominent intent features, and utilizes a bidirectional LSTM-based deep neural network for multi-label classification. We trained and evaluated \textsc{SmartIntentNN} on a dataset containing over 40,000 real-world smart contracts, employing self-comparison baselines in our experimental setup. The results show that \textsc{SmartIntentNN} achieves an F1-score of 0.8633 in identifying intents across 10 distinct categories, outperforming all baselines and addressing the gap in smart contract detection by incorporating intent analysis.
Jiho Choi, Junghoon Park, Woocheol Kim, Jin‐Hyeok Park · 6 authors
The recent advent of play-to-earn (P2E) systems in massively multiplayer online role-playing games (MMORPGs) has made in-game goods interchangeable with real-world values more than ever before. The goods in the P2E MMORPGs can be directly exchanged with cryptocurrencies such as Bitcoin, Ethereum, or Klaytn via blockchain networks. Unlike traditional in-game goods, once they had been written to the blockchains, P2E goods cannot be restored by the game operation teams even with chargeback fraud such as payment fraud, cancellation, or refund. To tackle the problem, we propose a novel chargeback fraud prediction method, PU GNN, which leverages graph attention networks with PU loss to capture both the players' in-game behavior with P2E token transaction patterns. With the adoption of modified GraphSMOTE, the proposed model handles the imbalanced distribution of labels in chargeback fraud datasets. The conducted experiments on three real-world P2E MMORPG datasets demonstrate that PU GNN achieves superior performances over previously suggested methods.
We investigate the use of deep learning to classify smart contract code vulnerabilities. We use different variants of Convolutional Neural Networks (CNNs) and a Long Short-Term Memory (LSTM) neural network. Five classes of vulnerabilities were employed. Our results suggest that the CNNs are able to provide a good level of accuracy, thus showing the viability of the proposed approach.
Blockchain technology has changed the global trading of assets. A blockchain can be viewed as a connected ledger managed by a distributed peer-topeer (P2P) network. Blockchain offers distinctive characteristics such as transactional privacy, the immutability of data, transparency and cryptographic, among others. These features paved the door for blockchain to develop numerous technology solution, including voting applications [1,2], internet of things (IoT) The increasing desire for technological advancements stimulated the development of BT.
With the development of Web 3.0 which emphasizes decentralization, blockchain technology ushers in its revolution and also brings numerous challenges, particularly in the field of cryptocurrency. Recently, a large number of criminal behaviors continuously emerge on blockchain, such as Ponzi schemes and phishing scams, which severely endanger decentralized finance. Existing graph-based abnormal behavior detection methods on blockchain usually focus on constructing homogeneous transaction graphs without distinguishing the heterogeneity of nodes and edges, resulting in partial loss of transaction pattern information. Although existing heterogeneous modeling methods can depict richer information through metapaths, the extracted metapaths generally neglect temporal dependencies between entities and do not reflect real behavior. In this paper, we introduce Time-aware Metapath Feature Augmentation (TMFAug) as a plug-and-play module to capture the real metapath-based transaction patterns during Ponzi scheme detection on Ethereum. The proposed module can be adaptively combined with existing graph-based Ponzi detection methods. Extensive experimental results show that our TMFAug can help existing Ponzi detection methods achieve significant performance improvements on the Ethereum dataset, indicating the effectiveness of heterogeneous temporal information for Ponzi scheme detection.