Yanan Gong, K. P. Chow, Siu Ming Yiu, Hing Fung Ting
Bitcoin is a widely used decentralized cryptocurrency. The proportion of Bitcoin transactions used for illegal activities is increasing. Mixing services are commonly applied to enhance anonymity and make transaction records more challenging to follow and analyze. The current research on peeling chains is generally based on heuristic algorithms to identify change addresses. However, due to the characteristics and limitations of the Bitcoin blockchain, there is no such ground truth to ensure the accuracy of each derived change address. This research analyzes the peeling chain patterns based on self-change addresses. The use of self-change addresses implies that the input address and the address used for receiving the change are controlled by the same entity. Also, each chain's transaction details and generated chain parameters are further verified for more precise results. Combining the two methods ensures the accuracy of the extracted peeling chains to some extent. And the corresponding behavior pattern of the extracted chains is studied.
Temporal Heterogeneous Networks (THNs) are evolving networks that characterize many real-world applications such as citation and events networks, recommender systems, and knowledge graphs. Forecasting THNs involves predicting future connections within a network that evolves over time and comprises diverse types of nodes and interactions with varying temporal dynamics. Although some Graph Neural Networks (GNNs) models have been successfully applied to forecast THNs, there is a lack of a general overview of how the message-passing computation could be extended to treat THNs. Moreover, most of the current solutions exhibit pitfalls in their training and evaluation strategies. Hence, in this work, we propose a graph deep learning framework for THN forecasting. Our framework decomposes the computation of a GNN layer into multiple components and introduces two different schemes to update embedding representations for THNs. This design allows the classification of existing solutions into special instances of our framework and highlights their potential limitations. We also extend the set of benchmarks for THNs by introducing two novel high-resolution temporal heterogeneous graph datasets derived from an emerging Web3 platform and a well-established e-commerce website. Overall, we conducted the first massive evaluation of THNs solutions over four temporal heterogeneous network datasets on two different future link prediction tasks using a fair newly introduced evaluation setting that considers the evolving nature of the data. Based on the limitations of existing solutions, we develop a new model that combines working techniques from previous models and leverages a new embedding update scheme. Experiments show the prediction power of our model compared to current solutions for link prediction in temporal graphs. Moreover, the experimental evaluation highlights the strengths and weaknesses of the different solutions and shows the effectiveness of our framework design.
Stefan Kambiz Behfar, Richard Mortier, Jon Crowcroft
The rapid growth of the Ethereum network necessitates advanced anomaly detection techniques to enhance security, transparency, and resilience against evolving malicious activities. While there have been significant strides in anomaly detection, they often fall short in capturing the intricate spatial-temporal patterns inherent in blockchain transactional data. This study presents a scalable framework that integrates Graph Convolutional Networks (GCNs) with Temporal Random Walks (TRW) specifically designed to adapt to the complexities and temporal dynamics of the Ethereum transaction network. Unlike traditional methods that focus on detecting specific attack types, such as front-running or flash loan exploits, our approach targets time-sensitive anomalies more broadlyâdetecting irregularities such as rapid transaction bursts, anomalous token swaps, and sudden volume spikes. This broader focus reduces reliance on pre-defined attack categories, making the method more adaptable to emerging and evolving malicious strategies. To ground our contributions, we establish three theoretical results: (1) the effectiveness of TRW in enhancing GCN-based anomaly detection by capturing temporal dependencies, (2) the identification of weight cancellation conditions in the anomaly detection process, and (3) the scalability and efficiency improvements of GCNs achieved through probabilistic sampling. Empirical evaluations demonstrate that the TRW-GCN framework outperforms state-of-the-art Temporal Graph Attention Networks (TGAT) in detecting time-sensitive anomalies. Furthermore, as part of our ablation study, we evaluated various anomaly detection techniques on the TRW-GCN embeddings and found that our proposed scoring classifier consistently achieves higher accuracy and precision compared to baseline methods such as Isolation Forest, One-Class SVM, and DBSCAN, thereby validating the robustness and adaptability of our framework.
Edgardo Brigatti, V. Rocha Grecco, Alexis HernĂĄndez, MĂĄrio Augusto Bertella
We introduce a general framework for empirically detecting interactions in communities of entities characterized by different features. This approach is inspired by ideas and methods coming from ecology and finance and is applied to a large dataset extracted from the cryptocurrency market. The inter-species interaction network is constructed using a similarity measure based on the log-growth rate of the capitalizations of the cryptocurrency market. The detected relevant interactions are only of the cooperative type, and the network presents a well-defined clustered structure, with two practically disjointed communities. The first one is made up of highly capitalized cryptocurrencies that are tightly connected, and the second one is made up of small-cap cryptocurrencies that are loosely linked. This approach based on the log-growth rate, instead of the conventional price returns, seems to enhance the discriminative potential of the network representation, highlighting a modular structure with compact communities and a rich hierarchy that can be ascribed to different functional groups. In fact, inside the community of the more capitalized coins, we can distinguish between clusters composed of some of the more popular first-generation cryptocurrencies, and clusters made up of second-generation cryptocurrencies. Alternatively, we construct the network of directed interactions by using the partial correlations of the log-growth rate. This network displays the important centrality of Bitcoin, discloses a core cluster containing a branch with the most capitalized first-generation cryptocurrencies, and emphasizes interesting correspondences between the detected direct pair interactions and specific features of the related currencies. As risk strongly depends on the interaction structure of the cryptocurrency system, these results can be useful for assisting in hedging risks. The inferred network topology suggests fewer probable widespread contagions. Moreover, as the riskier coins do not strongly interact with the others, it is more difficult that they can drive the market to more fragile states.
Hamzah M. Marhoon, Noorulden Basil, Alfian Maâarif
Blockchain technology is relatively young but has the potential to disrupt several industries. Since the emergence of Bitcoin, also known as Blockchain 1.0, there has been significant interest in this technology. The introduction of Ethereum, or Blockchain 2.0, has expanded the types of data that can be stored on blockchain networks. The increasing popularity of blockchain technology has given rise to new challenges, such as user privacy and illicit financial activities, but has also facilitated technical advancements. Blockchain technology utilizes cryptographic hashes of user input to record transactions. The public availability of blockchain data presents a unique opportunity for academics to analyze it and gain a better understanding of the challenges in blockchain communications. Researchers have never had access to such an opportunity before. Therefore, it is crucial to highlight the research problems, accomplishments, and potential trends and challenges in blockchain network data analysis and communications. This article aims to examine and summarize the field of blockchain data analysis and communications. The review encompasses the fundamental data types, analytical techniques, architecture, and operations related to blockchain networks. Seven research challenges are addressed: entity recognition, privacy, risk analysis, network visualization, network structure, market impact, and transaction pattern recognition. The latter half of this section discusses future research directions, opportunities, and challenges based on previous research limitations.
Attention-Aware Social Choice tackles the fundamental conflict faced by some agent communities between their desire to include all members in the decision making processes and the limited time and attention that are at the disposal of the community members. Here, we investigate a combination of two techniques for attention-aware social choice, namely Natural Language Processing (NLP) and Sampling. Essentially, we propose a system in which each governance proposal to change the status quo is first sent to a trained NLP model that estimates the probability that the proposal would pass if all community members directly vote on it; then, based on such an estimation, a population sample of a certain size is being selected and the proposal is decided upon by taking the sample majority. We develop several concrete algorithms following the scheme described above and evaluate them using various data, including such from several Decentralized Autonomous Organizations (DAOs).
Methodologies to infer financial networks from the price series of speculative assets vary, however, they generally involve bivariate or multivariate predictive modelling to reveal causal and correlational structures within the time series data. The required model complexity intimately relates to the underlying market efficiency, where one expects a highly developed and efficient market to display very few simple relationships in price data. This has spurred research into the applications of complex nonlinear models for developed markets. However, it remains unclear if simple models can provide meaningful and insightful descriptions of the dependency and interconnectedness of the rapidly developed cryptocurrency market. Here we show that multivariate linear models can create informative cryptocurrency networks that reflect economic intuition, and demonstrate the importance of high-influence nodes. The resulting network confirms that node degree, a measure of influence, is significantly correlated to the market capitalisation of each coin ($Ď=0.193$). However, there remains a proportion of nodes whose influence extends beyond what their market capitalisation would imply. We demonstrate that simple linear model structure reveals an inherent complexity associated with the interconnected nature of the data, supporting the use of multivariate modelling to prevent surrogate effects and achieve accurate causal representation. In a reductive experiment we show that most of the network structure is contained within a small portion of the network, consistent with the Pareto principle, whereby a fraction of the inputs generates a large proportion of the effects. Our results demonstrate that simple multivariate models provide nontrivial information about cryptocurrency market dynamics, and that these dynamics largely depend upon a few key high-influence coins.
This chapter highlights the need for a systemic, multi-disciplinary approach to understanding Web3's development. By disaggregating Web3 into interrelated networks of actors that hold distinct interests and beliefs, this chapter aims to describe and map Web3 to advance a more systemic understanding of its order and functioning. While presenting Web3 as a black box, the work underscores the importance of considering both the inherent nature of the technology itself and the broader societal context in which it operates and interacts with. Employing a Science, Technology, and Society lens, the work combines industry insights with a holistic approach to conceptualize the values, interests, and risks associated with Web3. The research holds that systemic challenges flow from Web3's inherent socio-technical and early-stage nature, and that its development is neither solely technologically determined nor entirely reliant on social actors. Instead, Web3's trajectory results from a complex interplay of its technological nature, societal values, stakeholder interests, and external (f)actors.
Abstract The growing attention on cryptocurrencies has led to increasing research on digital stock markets. Approaches and tools usually applied to characterize standard stocks have been applied to the digital ones. Among these tools is the identification of processes of market fluctuations. Being interesting stochastic processes, the usual statistical methods are appropriate tools for their reconstruction. There, besides chance, the description of a behavioural component shall be present whenever a deterministic pattern is ever found. Markov approaches are at the leading edge of this endeavour. In this paper, Markov chains of orders one to eight are considered as a way to forecast the dynamics of three major cryptocurrencies. It is accomplished using an empirical basis of intra-day returns. Besides forecasting, we investigate the existence of eventual long-memory components in each of those stochastic processes. Results show that predictions obtained from using the empirical probabilities are better than random choices.
The emergence of the Web3 paradigm has led to more and more systems built on blockchain technology and relying on cryptocurrency tokens â both fungible and non-fungible â to sustain themselves and generate profit. The growth and success of these platforms are strongly dependent on the growth and evolution of the trade relationships among users. In this context, it is of paramount importance to understand the mechanism behind the evolution and growth dynamics of these economic ties: however, in these systems the trade relationships are strictly intertwined with social dynamics, posing significant challenges in the analysis. One of the most important mechanisms behind the evolution of social networks is the triadic closure principle: given the strict link between social and economic spheres, the mechanism emerges as a potential candidate among mechanisms in literature. Therefore in this work, we extend the existing methodology for triadic closure studies and adapt it to directed networks. We performed an analysis centered around 3-node subgraphs known as âtriadsâ and statistically significant triads referred to as âtriadic motifsâ, both from a static and temporal perspective. The methodology was applied to various decentralized socio-economic networks with distinct levels of social components. These networks include currency transfers from the blockchain-based online social media platform Steemit, trade relationships among NFT sellers and buyers on the Ethereum blockchain, and a blockchain-based currency designed for humanitarian aid called Sarafu. Our measurements show how triadic closure is relevant during the evolution of these platforms and, for a few aspects, more impactful than centralized online social networks, where triadic closure is also incentivized by recommendation systems. Moreover, we are able to highlight both similarities and differences across networks with different levels of social components, both from a static and temporal standpoint. Overall our work presents strong evidence that triadic closure is an important evolutionary mechanism in decentralized socio-economic networks. Our findings provide a stepping stone in the study of decentralized socio-economic networks. Understanding the evolution of other decentralized networks, not following the same Web3 paradigm or with different social components will provide valuable insight into the understanding of dynamics in decentralized systems and potentially improve their design process.
This paper addresses the lack of research on quantifying Maximal Extractable Value (MEV) on Ethereum Layer 2 networks (L2s). Our findings reveal a substantial amount of MEV to be extracted on L2s, particularly on Polygon, with a lower bound of $213 million surpassing previous estimates. We observe that the majority of detected MEV on L2s consists of arbitrage opportunities, as liquidations are rare. These results emphasize the need for continuous monitoring and analysis of MEV on L2s, promoting informed decision-making for network selection and highlighting the associated risks.
Beibei Han, Yingmei Wei, Qingyong Wang, Francesco Maria De Collibus ¡ 5 authors
Abstract In recent years, a surge of criminal activities with cross-cryptocurrency trades have emerged in Ethereum, the second-largest public blockchain platform. Most of the existing anomaly detection methods utilize the traditional machine learning with feature engineering or graph representation learning technique to capture the information in transaction network. However, these methods either ignore the timestamp information and the transaction flow direction information in transaction network or only consider single transaction network, the cross-cryptocurrency trading patterns in Ethereum are usually ignored. In this paper, we introduce a Multi-layer Temporal Transaction Anomaly Detection (MT $$^2$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msup> <mml:mrow/> <mml:mn>2</mml:mn> </mml:msup> </mml:math> AD) model in Ethereum network with graph neural network. Specifically, for a given Ethereum token transaction network, we first extract its initial features including the structure subgraph and edgeâs feature. Then, we model the temporal information in subgraph as a series of network snapshots according to the timestamp on each edge and time window. To capture the cross-cryptocurrency trading patterns, we combine the snapshots from multiple token transactions at a given timestamp, and we consider it as a new combined graph. We further use the graph convolution encoder with attention mechanism and pooling operation on this new graph to obtain the graph-level embedding, and we transform the anomaly detection on dynamic multi-layer Ethereum transaction networks as a graph classification task with these graph-level embeddings. MT $$^2$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msup> <mml:mrow/> <mml:mn>2</mml:mn> </mml:msup> </mml:math> AD can integrate the transaction structure feature, edgeâs feature and cross-cryptocurrency trading patterns into a framework to perform the anomaly detection with graph neural networks. Experiments on three real-world multi-layer transaction networks show that the proposed MT $$^2$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msup> <mml:mrow/> <mml:mn>2</mml:mn> </mml:msup> </mml:math> AD (0.8789 Precision, 0.9375 Recall, 0.4987 FbMacro and 0.9351 FbWeighted) can achieve the best performance on most evaluation metrics in comparison with some competing approaches, and the effectiveness in consideration of multiple tokens is also demonstrated.
Nils Augustin, Andreas Eckhardt, Alexander Willem de Jong
Abstract Blockchain technology is argued to drastically change the way we operate within an organizational context, with decentralized autonomous organizations (DAOs) representing a first manifestation of this ongoing trend. DAOs are characterized by an online community that builds the organizationâs backbone by providing knowledge and human resources in a transparent, virtual manner, as well as the use of blockchain technology to coordinate their endeavor. Nevertheless, current research highlights the conceptual ambiguity of this emerging phenomenon, leading to potential issues for practitioners and researchers. To provide further clarity on the phenomenon, we study DAOs through the perspective of their members with a two-staged approach by combining elements of a netnographic approach and structural topic modeling. Our findings highlight several contextual features surrounding DAOs, such as their membersâ underlying beliefs and views, helping to embed DAOs in existing research streams.
In recent years, the rapid growth of blockchain technology has sparked massive curiosity and transformed various industries. Among the numerous blockchain platforms, Ethereum has gathered significant attention for its decentralized applications and smart contracts. Understanding Ethereum and its network interactions is a challenging task but with various methods at our disposal, such as graph embeddings, we gain valuable insight into its operations. Graph embeddings are powerful techniques in the realm of data representation, which have become a focal point in analyzing complex structures. By capturing the essence of graphâs structure and semantics, embeddings enable efficient analysis of vast networks. In the context of blockchain-induced networks, graph embeddings offer effective insights into the behavior and dynamics of transactions and addresses. In my Bachelorâs thesis, conducted under the guidance and support of Dr. Ferenc Beres and Marcell Nagy, I aim to explore the connection between graph embeddings and blockchain-induced networks from Ethereum with a binary classification problem on certain Ethereum accountsâ network interactions as graphs. Additionally, I analyze graph-level properties and employ dimensionality reduction techniques to visualize these networks in the embedding space.
Community detection is widely used in social networks to uncover groups of related vertices (nodes). In cryptocurrency transaction networks, community detection can help identify users that are most related to known illegal users. However, there are challenges in applying community detection in cryptocurrency transaction networks: (1) the use of pseudonymous addresses that are not directly linked to personal information make it difficult to interpret the detected communities; (2) on Bitcoin, a user usually owns multiple Bitcoin addresses, and nodes in transaction networks do not always represent users. Existing works on cluster analysis on Bitcoin transaction networks focus on addressing the later using different heuristics to cluster addresses that are controlled by the same user. This research focuses on illegal community detection containing one or more illegal Bitcoin addresses. We first investigate the structure of Bitcoin transaction networks and suitable community detection methods, then collect a set of illegal addresses and use them to label the detected communities. The results show that 0.06% of communities from daily transaction networks contain one or more illegal addresses when 2,313,344 illegal addresses are used to label the communities. The results also show that distance-based clustering methods and other methods depending on them, such as network representation learning, are not suitable for Bitcoin transaction networks while community quality optimization and label-propagation-based methods are the most suitable.
Nicolò Vallarano, Tiziano Squartini, Claudio J. Tessone
The open availability of the entire history of the Bitcoin transactions opens up the possibility to study this system at an unprecedented level of detail. This contribution is devoted to the analysis of the mesoscale structural properties of the Bitcoin User Network (BUN), across its entire history (i.e. from 2009 to 2017). What emerges from our analysis is that the BUN is characterized by a core-periphery structure a deeper analysis of which reveals a certain degree of bow-tieness (i.e. the presence of a Strongly-Connected Component, an IN- and an OUT-component together with some tendrils attached to the IN-component). Interestingly, the evolution of the BUN structural organization experiences fluctuations that seem to be correlated with the presence of bubbles, i.e. periods of price surge and decline observed throughout the entire Bitcoin history: our results, thus, further confirm the interplay between structural quantities and price movements observed in previous analyses.
Because of its ability to completely revamp current blockchain security methods, this connection is crucial. An effective safeguard against complex assaults, Nonlinear Dynamics (ND) adds a living, breathing component to consensus methods and cryptographic primitives. There is an urgent need for creative, nonlinear methods to strengthen blockchain security in light of present challenges including increasing attack vectors and risks posed by quantum computing. The suggested Dynamic Chaos-based Blockchain Security (DC-BS) system in this paper makes use of the chaotic dynamics present in ND to strengthen various aspects of blockchain security. Adaptive threat detection systems, dynamic consensus methods, and chaos-based encryption are all newly introduced in DC-BS. Validation of DC-BS's efficacy in preventing various attack scenarios through simulation studies demonstrates its advantages in reducing vulnerabilities and responding to new attack types. Various decentralized systems can benefit from DC-BS, including as supply chain management, the Internet of Things (IoT), conventional blockchain networks, and decentralized finance (DeFi). To strengthen the security of various decentralized applications, DC-BS works to increase trust, transparency, and resilience. The effectiveness of DCBS is confirmed by thorough simulation analyses that cover a wide range of attack scenarios, including double-spending assaults, Sybil attacks, and eclipse attacks. Based on the results of the simulations, DCBS is much more effective than conventional blockchain security procedures at reducing these risks. Showcased as well is the technique's capacity to react to changing attack techniques, highlighting its capacity to provide strong security even in dynamic settings.
Over the last fifteen years, the development of blockchain technologies has attracted a large volume of professional expertise, capital investment and media attention. This burgeoning sector of technology practices has coalesced around a few major initiatives (Bitcoin, Ethereum), but it is still moving at a fast pace and its configuration is evolving. If this sector is marked by a variety of technological protocols, financial arrangements and organizational forms, it is also, we would argue, a site of social effervescence. Parties, meet-ups, and the sorts of informal socializing which gather around events and networks of all kinds function to endow the blockchain sector with the characteristics of what, in cultural analysis, are often called âscenesâ. The aim of this special issue is to examine the interest of the notion of scene for the analysis of blockchain practices. We argue that the notion of scene may be mobilized as a useful analytical framework not only for the study of blockchain practices, but for that of technology practices more generally. In this introductory article, we ask the following questions: how can the notion of scene contribute to the understanding of blockchain practices? And what sort of research agenda does the notion point to? In the following sections we first identify some âscenicâ components in blockchain phenomena. Then we review how media discourses and academic scholarship have framed these phenomena to show that the scene perspective is undertheorized in the context of technology-related social groupings. Finally, we propose a framework to analyse the main dimensions of blockchain scenes, before presenting the contributions to the special issue. With this special issue, we aim to establish a research agenda around technology scenes at the junction of STS and cultural analysis.
Investigating the timeâfrequency-based phase synchronization between nonstationary time series of the cryptocurrenciesâ prices can be a suitable tool to reveal their complicated interactions at different frequencies. In this work, the phase synchronization between 25 cryptocurrencies with the highest capitalization from December 1, 2021 to June 1, 2022 is calculated using the phase-locking value method based on wavelet transform. Then, utilizing the graph theory, the cryptocurrency networks are constructed, and their topological features like path length (PL), clustering coefficient (CC) and node strength are evaluated in various frequencies. This research indicates a strong phase synchronization between the investigated cryptocurrencies, especially in low frequencies. Also, the networksâ high average CC and short PL compared to their equivalent regular and random networks display the small-worldness of the networks. We observe from the obtained results that Bitcoin, Ethereum and Binance currencies, among the most popular cryptocurrencies, have the highest average node strengths at different frequencies. Also, TRON shows the lowest CC and node strength among all currencies, representing its limited phase interaction with other currencies.
Blockchain has facilitated the growth of cryptocurrencies but has also provided new ideas for illegals to commit fraud. Research on malicious accounts detection shows that the number of malicious accounts is much smaller than that of benign accounts, leading to imbalanced dataset samples. Most researchers adopt the under-sampling method to help deal with this issue, but this method does not correspond to the actual scale. So, we propose an anomaly detection method based on community discovery. Firstly, we use the transaction information in the Ethereum public chain to build a transaction network and use the Louvain algorithm to divide the transaction network into communities. Secondly, we use the LightGBM algorithm to classify the community. Finally, based on the classification results, we use HBOS, LOF, K-Means, KNN and iForest algorithms as benchmark algorithms for anomaly detection and compare the experimental results using the methods in this paper with the results of anomaly detection using the original transaction network. Experimental show that our method can reduce the amount of data by 35.53% and increase the AUC values of the five algorithms by 7.52%, 8.41%, 14.88%, 0.83% and 27.95%.
The current online social network landscape is characterized by competition to get larger audiences leading to massive user migrations which will determine the shape of the future Web. However, user migration phenomena have not been fully understood and their driving mechanisms are still not well identified; in particular, the behaviors of hubs and the influence they exert on their followers are unclear. In this work, we focus on these aspects by analyzing the propensity of hubs to migrate towards a new social platform as a consequence of a shocking event; and the influence they exert on the decision of their neighbors of migrating to a new platform or staying on the native one. We conducted analysis on data made available after a user migration consequence of a hard fork involving two Web3 online social networks based on the blockchains Steem and Hive. Due to the blockchain nature of these Web3 platforms, we got detailed data about social and financial interactions among the users, along with information that allowed a precise reconstruction of the context surrounding the migration. The main findings suggest that different types of hubs apply different strategies when choosing to migrate, e.g. financial hubs diversify their strategy by staying and migrating at the same time. As for hub influence, results suggest that users directly interacting with hubs tend to migrate. In general, findings on influence indicate that understanding the activity and the influence of hubs is crucial in monitoring and controlling the user migration process.