Fog computing extends cloud computing services closer to users, improving efficiency and reducing latency. Smart contracts play a key role in authentication and resource access management within this framework. As the adoption of smart contracts in fog computing grows, ensuring their security has become a major challenge. This study enhances smart contract attack detection in fog computing using machine learning techniques. A dataset of 818 smart contracts was collected from “etherscan.io.” Feature extraction was performed using Word2Vec and BERT, while feature selection was done using the information gain method. The Random Forest (RF) and Extra Trees Classifier (ETC) achieved the highest accuracy of 0.91 with Word2Vec, while the LightGBM (LGBM) classifier attained 0.90 accuracy using BERT. These results demonstrate the effectiveness of machine learning models in improving smart contract security within fog computing environments.
The paper explores the revolutionary potential of Artificial Intelligence (AI) in Indonesia's financial ecosystem, highlighting its capacity to improve operational efficiency, foster financial inclusion, and tackle specific socio-economic concerns. This study emphasizes Indonesia's varied demographic and digital environment, illustrating how AI-driven innovations like decentralized finance (DeFi), predictive analytics, and blockchain integration transform financial products to cater to disadvantaged people. This study utilizes over 20 scholarly publications and international case studies to highlight the strategic significance of promoting ethical AI practices, mitigating algorithmic bias, and closing infrastructural and talent disparities to achieve sustainable and inclusive economic growth. The results support implementable methods, such as public-private collaborations, strong regulatory structures, and AI-driven individualized financial solutions, to optimize the advantages of digital transformation in Indonesia's financial industry. Future research must emphasize empirical investigations into AI's capacity to mitigate financial inequalities and stimulate regional innovation, thereby establishing Indonesia as a frontrunner in AI-facilitated economic transformation.
Stefan Dziembowski, Shahriar Ebrahimi, Parisa Hassanizadeh
Ensuring the authenticity and credibility of daily media on internet is an ongoing problem. Meanwhile, genuinely captured images often require refinements before publication. Zero-knowledge proofs (ZKPs) offer a solution by verifying edited image without disclosing the original source. However, ZKPs typically come with high costs, particularly in terms of prover complexity and proof size. This paper presents VIMz, a framework for efficiently proving the authenticity of high-resolution images using folding-based zkSNARKs; a type of proving system that minimizes computational overhead by recursively folding multiple evaluations of the same constraints into a compact proof. As a complete proof system, VIMz proves the integrity of both the original and edited images, as well as the correctness of the transformation without revealing intermediate images within a chain of edits--only the final result is disclosed. Moreover, VIMz maintains the anonymity of the original signer and all subsequent editors while proving the authenticity of the final image. We also compare VIMz with the system model in Coalition for Content Provenance and Authenticity (C2PA) from different perspectives and show that VIMz offers higher level of security guarantee by eliminating the need to trust the editing environment. Experimental results show that VIMz performs efficiently in both prover and verifier sides. It can prove the transformations on 8K (33MP,i.e., 100MB) images with up to 13%~25% faster than the competition, while reaching to a peak memory of only 10 GB. Moreover, VIMz has a verification time of under 1 second and achieves succinct proofs of less than 11 KB for all resolutions, which is more than 90% improvement compared to the competition. VIMz's low memory complexity allows for proving multiple transformations in parallel to achieve a 3.5x additional speedup on average.
Open access
Advanced Steganography and Watermarking Techniques
Digital Media Forensic Detection
Physical Unclonable Functions (PUFs) and Hardware Security
This work presents a mathematical solution to data privacy and integrity issues in Split Learning which uses Homomorphic Encryption (HE) and Zero-Knowledge Proofs (ZKP). It allows calculations to be conducted on encrypted data, keeping the data private, while ZKP ensures the correctness of these calculations without revealing the underlying data. Our proposed system, HavenSL, combines HE and ZKP to provide strong protection against attacks. It uses Discrete Cosine Transform (DCT) to analyze model updates in the frequency domain to detect unusual changes in parameters. HavenSL also has a rollback feature that brings the system back to a verified state if harmful changes are detected. Experiments on CIFAR-10, MNIST, and Fashion-MNIST datasets show that using Homomorphic Encryption and Zero-Knowledge Proofs during training is feasible and accuracy is maintained. This mathematical-based approach shows how crypto-graphic can protect decentralized learning systems. It also proves the practical use of HE and ZKP in secure, privacy-aware collaborative AI.
The increasing adoption of blockchain technology has led to a growing demand for higher transaction throughput. Traditional blockchain platforms, such as Ethereum, execute transactions sequentially within each block, limiting scalability. Parallel execution has been proposed to enhance performance, but existing approaches either impose strict dependency annotations, rely on conservative static analysis, or suffer from high contention due to inefficient state management. Moreover, even when transaction execution is parallelized at the upper layer, storage operations remain a bottleneck due to sequential state access and I/O amplification. In this paper, we propose Reddio, a batch-based parallel transaction execution framework with asynchronous storage. Reddio processes transactions in parallel while addressing the storage bottleneck through three key techniques: (i) direct state reading, which enables efficient state access without traversing the Merkle Patricia Trie (MPT); (ii) asynchronous parallel node loading, which preloads trie nodes concurrently with execution to reduce I/O overhead; and (iii) pipelined workflow, which decouples execution, state reading, and storage updates into overlapping phases to maximize hardware utilization.
Minh Trung Tran, Nasrin Sohrabi, Zahir Tari, Qin Wang · 6 authors
We identify the slow liquidity drain (SLID) scam, an insidious and highly profitable threat to decentralized finance (DeFi), posing a large-scale, persistent, and growing risk to the ecosystem. Unlike traditional scams such as rug pulls or honeypots (USENIX Sec'19, USENIX Sec'23), SLID gradually siphons funds from liquidity pools over extended periods, making detection significantly more challenging. In this paper, we conducted the first large-scale empirical analysis of 319,166 liquidity pools across six major decentralized exchanges (DEXs) since 2018. We identified 3,117 SLID affected liquidity pools, resulting in cumulative losses of more than US$103 million. We propose a rule-based heuristic and an enhanced machine learning model for early detection. Our machine learning model achieves a detection speed 4.77 times faster than the heuristic while maintaining 95% accuracy. Our study establishes a foundation for protecting DeFi investors at an early stage and promoting transparency in the DeFi ecosystem.
Cláudio Massingarela, Rabeca Cuna, Alberto Daniel, Filipe Mahaluça
This study examines the viability of Bitcoin investment in Maputo, Mozambique, within the context of the growing global adoption of cryptocurrencies. The research highlights the advantages of Bitcoin, such as lower transaction fees and the elimination of financial intermediaries, offering greater efficiency and flexibility for merchants and investors. The blockchain technology underlying Bitcoin provides security and privacy in transactions, making them resistant to fraud. However, Bitcoin's high volatility presents a significant challenge, particularly in unregulated markets like Mozambique, where the lack of a clear regulatory framework limits wider adoption. Using a mixed-method approach, quantitative data were collected from 23 Bitcoin investors through structured questionnaires, and qualitative data were gathered from semi-structured interviews with three investors and a financial analyst. Statistical analysis, conducted using R software, included tests such as Chi-Square, Student’s t-test, Mann-Whitney U, Pearson correlation, logistic regression, and factor analysis to understand the investment patterns and motivations of investors. The results revealed no significant association between gender and Bitcoin investment recommendations, although education level showed a marginally significant relationship, indicating that individuals with higher financial literacy are more likely to recommend Bitcoin. The analysis also found no significant differences in investment returns between men and women, suggesting that investment strategy plays a more crucial role. Risk-seeking investors achieved substantially higher returns, reflecting the speculative nature of Bitcoin. The study's limitations include the small sample size and the lack of specific cryptocurrency regulations in Mozambique, which restrict the generalizability of the findings. Recommendations include promoting financial education programs on cryptocurrencies, considering the legalization and regulation of Bitcoin by the Central Bank of Mozambique, increasing the banking sector's involvement in cryptocurrency discussions, and expanding research on Bitcoin volatility and returns. These actions could contribute to a more secure and informed investment environment in Mozambique.
In their recent breakthrough result, Slofstra and the second author show that there is a two-player one-round perfect zero-knowledge MIP* protocol for RE (STOC'24). We build on their result to show that there exists a succinct two-player one-round perfect zero-knowledge MIP* protocol for RE against dishonest verifiers with polylog question size and O(1) answer size, or with O(1) question size and polylog answer size. To prove our result, we study the three central compression techniques underlying the MIP*=RE proof (Ji et al. '20): question reduction, oracularization, and answer reduction. We show that question reduction preserves the perfect (as well as statistical and computational) zero-knowledge properties of the original protocol against dishonest verifiers, and oracularization and answer reduction preserve the perfect (as well as statistical and computational) zero-knowledge properties of the original protocol against honest verifiers. Secondly, we show that every constraint-constraint binary constraint system (BCS) nonlocal game, which provides a quantum information characterization of MIP*, can be converted to a synchronous constraint-variable BCS game to preserve perfect completeness for our compression. Lastly, we present a parametrized perfect-zero-knowledge transformation of MIP* protocols, which generalizes the transformation in (Slofstra and Kieran STOC'24) . This transformation allows us to preserve the zero-knowledge property against dishonest verifiers in the recursively oracularized protocols in our compression.
We analyze a model of heterogeneous rational bubbles that compete and complement each other. When some bubbles burst, surviving ones gain value, offsetting losses from collapsed bubbles. This “compensation effect,” combined with diversification, enhances welfare. A portfolio of fragile bubbles may rival a single, stable bubble. The stationary equilibrium imposes a tight upper bound on bubble size, considering covariance structures, price fluctuations, and the emergence of new bubbles. These results have important policy implications, particularly for managing crypto ETFs and issuing CBDCs, highlighting the potential benefits of a diversified approach to fragile financial systems. • We study a model of heterogeneous rational bubbles that compete and complement each other. • A bubble’s market size is driven by agents’ confidence, with greater confidence leading to larger bubbles. • When some bubbles burst, survivors appreciate in value, offsetting losses and mitigating welfare impacts. • A diversified portfolio of fragile bubbles such as a crypto ETF may rival a single, stable bubble thanks to this “compensation effect”.
With the growing demand for secure, decentralized file sharing solutions, this study presents a blockchain and IPFS-based framework for efficient data storage and access control. The proposed system leverages asymmetric encryption, smart contracts, and distributed access management to ensure confidentiality and integrity. The files are encrypted using AES-256 before they are stored in IPFS, and SHA-256 hashing is used to verify the content. Access control is guaranteed using blockchain-based policies, encrypting access keys and dynamic permissions to ensure that users can exchange files. An intelligent contract automates authentication, access and distribution of keys, minimizing dependence on centralized bodies. In addition, suppliers of storage facilities for an incentive mechanism reward the economy of economic and scalable storage. By using consensus mechanisms, such as proof of aspiration (POS) or proof of the authorities (POA), the system prevents unauthorized modifications and increases data security. This approach provides a reliable solution for organizations requiring controlled access to confidential data, with potential applications in the field of financial, health care and public sectors.
In recent years, the convergence between blockchain and artificial intelligence (AI) has led to significant innovations in the agricultural sector, particularly in the traceability and protection of grains. These emerging technologies have the potential to transform the agricultural supply chain, providing greater transparency, security, and efficiency. Blockchain technology, with its ability to create immutable and transparent records, is widely applied to trace the origin and movement of grains from production to the final consumer. At the same time, AI plays a key role in analyzing large volumes of data, allowing for the prediction of risks and the dynamic adaptation of agricultural insurance contracts. Additionally, the combination of blockchain and AI facilitates the creation of new financing models, such as smart contracts, which automatically execute when certain conditions are met. These advancements help ensure the quality of grains, combat fraud, optimize logistics processes, and respond more swiftly to unforeseen events. The integration of these technologies also contributes to more sustainable, efficient, and resilient agriculture, addressing challenges such as climate change, price volatility, and the increasing demand for transparency in the supply chain. The combined use of blockchain and AI is reshaping grain production and traceability, providing a safer and more efficient system for the future of agriculture, particularly in the United States.
Lucien K. L. Ng, Pedro Moreno-Sánchez, Mohsen Minaei, Panagiotis Chatzigiannis · 6 authors
Zero-Knowledge Succinct Non-Interactive Argument of Knowledge (zk-SNARK) schemes have gained significant adoption in privacy-preserving applications, decentralized systems (e.g., blockchain), and verifiable computation due to their efficiency. However, the most efficient zk-SNARKs often rely on a one-time trusted setup to generate a public parameter, often known as the ``Powers of Tau" (PoT) string. The leakage of the secret parameter, $τ$, in the string would allow attackers to generate false proofs, compromising the soundness of all zk-SNARK systems built on it. Prior proposals for decentralized setup ceremonies have utilized blockchain-based smart contracts to allow any party to contribute randomness to $τ$ while also preventing censorship of contributions. For a PoT string of $d$-degree generated by the randomness of $m$ contributors, these solutions required a total of $O(md)$ on-chain operations (i.e., in terms of both storage and cryptographic operations). These operations primarily consisted of costly group operations, particularly scalar multiplication on pairing curves, which discouraged participation and limited the impact of decentralization In this work, we present Lite-PoT, which includes two key protocols designed to reduce participation costs: \emph{(i)} a fraud-proof protocol to reduce the number of expensive on-chain cryptographic group operations to $O(1)$ per contributor. Our experimental results show that (with one transaction per update) our protocol enables decentralized ceremonies for PoT strings up to a $2^{15}$ degree, an $\approx 16x$ improvement over existing on-chain solutions; \emph{(ii)} a proof aggregation technique that batches $m$ randomness contributions into one on-chain update with only $O(d)$ on-chain operations, independent of $m$. This significantly reduces the monetary cost of on-chain updates by $m$-fold via amortization.
We propose Data Tumbling Layer (DTL), a cryptographic scheme for non-interactive data tumbling. The core concept is to enable users to commit to specific data and subsequently re-use to the encrypted version of these data across different applications while removing the link to the previous data commit action. We define the following security and privacy notions for DTL: (i) no one-more redemption: a malicious user cannot redeem and use the same data more than the number of times they have committed the data; (ii) theft prevention: a malicious user cannot use data that has not been committed by them; (iii) non-slanderabilty: a malicious user cannot prevent an honest user from using their previously committed data; and (iv) unlinkability: a malicious user cannot link tainted data from an honest user to the corresponding data after it has been tumbled. To showcase the practicality of DTL, we use DTL to realize applications for (a) unlinkable fixed-amount payments; (b) unlinkable and confidential payments for variable amounts; (c) unlinkable weighted voting protocol. Finally, we implemented and evaluated all the proposed applications. For the unlinkable and confidential payment application, a user can initiate such a transaction in less than $1.5$s on a personal laptop. In terms of on-chain verification, the gas cost is less than $1.8$ million.
Understanding how central and local governments share resources and responsibilities is crucial for analyzing political and economic systems.Decentralization is not a one-size-fits-all solution for enhancing local government efficiency and responsiveness.While it was once believed to lead to better governance and civic engagement, fiscal challenges (such as vertical fiscal imbalances, soft budget constraints, and the flypaper effect) can undermine fiscal discipline and efficiency, potentially causing fiscal crises at the subnational level.This thesis examines fiscal decentralization and public finance in Brazil through three empirical essays.First, I explore the financial impacts of extreme weather events on local public finances in Brazil.The findings show that droughts do not significantly influence intergovernmental transfers, causing financial strain, while floods result in increased government grants.However, this financial boost does not lead to better spending on flood mitigation, indicating a moral hazard associated with reliance on higher-level government resources.Second, I investigate the impact of territorial divisions on local governments.The analysis, using voter turnout and financial data, shows that administrative divisions initially boost electoral engagement, though this effect fades over time.Territorial fragmentation also increases reliance on vertical transfers while raising expenditures without significantly affecting fiscal balance.Third, I evaluate the Program for the Modernization of Tax Administration (PMAT), which was designed to enhance local tax collection.This analysis shows that the program had no significant impact on tax collection, highlighting the ineffectiveness of modernization efforts aimed at reducing municipal reliance on intergovernmental transfers.
Yuming Huang, Jing Tang, Qianhao Cong, T. B. Richard · 6 authors
In blockchains using the Proof-of-Work (PoW) consensus mechanism, a mining pool is a joint group of miners who combine their computational resources and share the generated revenue. Similarly, when the Proof-of-Stake (PoS) consensus mechanism is adopted, the staking pool imitates the design of the mining pool by aggregating the stakes. However, in PoW blockchains, the pooling approach has been criticized to be vulnerable to the block withholding (BWH) attack. BWH attackers may steal the dividends from victims by pretending to work but making invalid contributions to the victim pools. It is well known that BWH attackers against PoW face the miner's dilemma . To our knowledge, despite the popularity of PoS, we are the first to study the pool BWH attack against PoS. Interestingly, we find that, for a network only consisting of one attacker pool and one victim pool, the attacker will eventually manipulate the network while the victim will vanish by losing the stake ratio gradually. Moreover, in a more realistic scenario with multiple BWH attacker pools and one solo staker who does not join any pools, we show that only one lucky attacker and the solo staker will survive, whereas all the other pools will vanish gradually, revealing the staker's dilemma . These findings indicate that, compared to PoW, the BWH attack on PoS has a much more severe impact due to the attacker's resource aggregation advantage. Our analysis is supported by experiments on massive real blockchain systems and numerical simulations.
М. А. Абрамова, С. В. Криворучко, Oleg V. Lunyakov, Алим Борисович Фиапшев
Existing studies of the problem of the emergence and development of decentralized finance (DeFi) are largely limited to non-principled clarification of certain positions and formulations, with emphasis on technical and technological innovations, far from the level of fundamental research. The authors set the task of theoretical understanding of the ongoing transformation processes in the financial sphere. The purpose of the study was to identify the conditions, driving forces and nature of the process of development of decentralized finance; to define DeFi and identify its sustainable features; and to substantiate the possibilities of considering DeFi as a separate economic category and institution. Setting the goal determined the sequence of its solution in two stages. The first stage implied a higher level of abstraction, an appeal to the theory of money and its modern achievements. The second stage —“movement to the surface”, inclusion in the analysis of specifications accompanying the development of DeFi. The authors used systematic and logical methods , induction and deduction as the main methods, which allowed them to generalize and systematize the ideas about the essence of decentralized finance, identify problems in the modern scientific discourse. As a result, the causes are revealed, and the nature of the process of emergence and development of the sphere of decentralized finance is substantiated, the definition of DeFi is given, the principles of their functioning are highlighted and recommendations on structuring the conceptual apparatus of DeFi are developed. It is concluded that the process of formation and development of decentralized finance is objective and driven by changes in the monetary sphere, technological advances, and problems of traditional finance. At the same time, the stable features of DeFi determine the potential of reproduction of financial relations on a decentralized basis, but at the same time do not allow us to qualify DeFi as an independent category and institution. The results of the study can be used both in elaborating the concept of DeFi development and taken into account as part of the regulatory response to DeFi.
In a world where data is the new currency, wearable health devices offer unprecedented insights into daily life, continuously monitoring vital signs and metrics. However, this convenience raises privacy concerns, as these devices collect sensitive data that can be misused or breached. Traditional measures often fail due to real-time data processing needs and limited device power. Users also lack awareness and control over data sharing and usage. We propose a Privacy-Enhancing Technology (PET) framework for wearable devices, integrating federated learning, lightweight cryptographic methods, and selectively deployed blockchain technology. The blockchain acts as a secure ledger triggered only upon data transfer requests, granting users real-time notifications and control. By dismantling data monopolies, this approach returns data sovereignty to individuals. Through real-world applications like secure medical data sharing, privacy-preserving fitness tracking, and continuous health monitoring, our framework reduces privacy risks by up to 70 percent while preserving data utility and performance. This innovation sets a new benchmark for wearable privacy and can scale to broader IoT ecosystems, including smart homes and industry. As data continues to shape our digital landscape, our research underscores the critical need to maintain privacy and user control at the forefront of technological progress.
Parallel execution of smart contract transactions in large multicore architectures is critical for higher efficiency and improved throughput. The main bottleneck for maximizing the throughput of a node through parallel execution is transaction conflict resolution: when two transactions interact with the same data, like an account balance, their order matters. Imagine one transaction sends tokens from account A to account B, and another tries to send tokens from account B to account C. If the second transaction happens before the first one, the token balance in account B might be wrong, causing the entire system to break. Conflicts like these must be managed carefully, or you end up with an inconsistent, unusable blockchain state. Traditional software transactional memory (STM) has been identified as a possible abstraction for the concurrent execution of transactions within a block, with Block-STM pioneering its application for efficient blockchain transaction processing on multicore validator nodes. This paper presents a parallel execution methodology that leverages conflict specification information of the transactions for block transactional memory (BTM) algorithms. Our experimental analysis, conducted over synthetic transactional workloads and real-world blocks, demonstrates that BTMs leveraging conflict specifications outperform their plain counterparts on both EVM and MoveVM. Our proposed BTM implementations achieve up to 1.75x speedup over sequential execution and outperform the state-of-the-art Parallel-EVM (PEVM) execution by up to 1.33x across synthetic workloads.
Fei Ren, Bo Zhao, Jun Wang, Juxiang Zhou · 5 authors
With the rapid development of information technology, blended learning has become a crucial aspect of modern education. However, the fragmented use of various teaching platforms, such as Xuexitong and Rain Classroom, has led to the dispersion of teaching data. This not only increases the cognitive load on teachers and students but also hinders the systematic recording of teaching activities and learning outcomes. Moreover, existing blended learning evaluation systems exhibit significant shortcomings in large-scale data storage and secure sharing. To address these issues, this study designs a blended teaching evaluation management system based on blockchain and searchable encryption. First, an on-chain and off-chain collaborative storage model is established using the Ethereum blockchain and the InterPlanetary File System (IPFS) to ensure secure and large-scale storage of student work data. Next, a role-based access control scheme utilizing smart contracts is proposed to effectively prevent unauthorized access. Simultaneously, a searchable encryption scheme is designed using AES-CBC-256 and SHA-256 algorithms, enabling data sharing while safeguarding data privacy. Additionally, the smart contract comprehensively records students’ grade information, including weekly regular scores, midterm scores, final scores, overall scores, and their rankings, ensuring transparency in the evaluation process. Based on these technical solutions, a general-purpose teaching evaluation management system (B-Education) is developed. The experimental results demonstrate that the system accurately records teaching activities and learning outcomes, improving the transparency of teaching evaluations while ensuring data security and privacy. The system’s gas consumption remains within a reasonable range, demonstrating good flexibility and usability. Educational institutions can flexibly configure course evaluation criteria and adjust the weighting of various grades based on their specific needs. This study provides an innovative solution for blended teaching evaluation, offering significant theoretical value and practical implications.
Blockchain, with its features of decentralization, immutability and security, has emerged as a revolutionary technology. However, the rapid growth of distributed ledgers leads to escalating storage costs, high transaction verification overhead and low throughput, limiting its scalability in broader applications. To address these challenges, this study proposes a novel blockchain transaction verification scheme, BBVC, specifically designed for smart contracts. BBVC leverages a vector commitment algorithm to generate transaction commitments and introduces a two-layer aggregation technique to compress node state data, facilitating efficient transaction validity proof aggregation and verification. A versioning mechanism is further incorporated to optimize proof update efficiency to sub-linear levels, while a new contract transaction verification logic accelerates transaction processing without compromising security. Experimental results demonstrate that the BBVC scheme maintains a constant proof size of 48 bytes and achieves millisecond-level proof generation and verification for large-scale transactions. Compared to Hyperproofs, BBVC enhances transaction aggregation efficiency by 250 times and verification efficiency by 22.5–27.5 times, significantly advancing blockchain scalability and performance.
This comprehensive article explores recent advancements in privacy-preserving technologies within artificial intelligence systems, focusing on five key approaches: federated learning, differential privacy, homomorphic encryption, privacy-preserving machine learning (PPML), and zero-knowledge proofs. The article examines how these technologies address critical privacy challenges in machine learning environments while maintaining model performance and utility. The article highlights the implementation of these approaches across various domains, particularly in healthcare and financial services, demonstrating their effectiveness in protecting sensitive data throughout the machine learning lifecycle. The article reveals how these technologies complement each other to create robust privacy protection frameworks while enabling organizations to leverage the power of AI without compromising data confidentiality.
Abstract Federated Learning (FL) is a promising form of distributed machine learning that preserves privacy by training models locally without sharing raw data. While FL ensures data privacy through collaborative learning, it faces several critical challenges. These include vulnerabilities to reverse engineering, risks to model architecture privacy, susceptibility to model poisoning attacks, threats to data integrity, and the high costs associated with communication and connectivity. This paper presents a comprehensive review of FL, categorizing data partitioning formats into horizontal federated learning, vertical federated learning, and federated transfer learning. Furthermore, it explores the integration of FL with blockchain, leveraging blockchain’s decentralized nature to enhance FL’s security, reliability, and performance. The study reviews existing FL models, identifying key challenges such as privacy risks, communication overhead, model poisoning vulnerabilities, and ethical dilemmas. It evaluates privacy-preserving mechanisms and security strategies in FL, particularly those enabled by blockchain, such as cryptographic methods, decentralized consensus protocols, and tamper-proof data logging. Additionally, the research analyzes regulatory and ethical considerations for adopting blockchain-based FL solutions. Key findings highlight the effectiveness of blockchain in addressing FL challenges, particularly in mitigating model poisoning, ensuring data integrity, and reducing communication costs. The paper concludes with future directions for integrating blockchain and FL, emphasizing areas such as interoperability, lightweight consensus mechanisms, and regulatory compliance.
Introduzione: I processi di digitalizzazione nelle attività creative permettono di ottenere nuovi contenuti attraverso l'uso di dati e algoritmi di machine learning, creando relazioni inedite. In questo contesto, il diritto d’autore deve proteggere gli autori senza ostacolare l'uso dei dati virtuali, necessari per risultati originali. Metodologia: La ricerca esplora l'impatto dei non-fungible token (NFT), tecnologia emergente che ha rivoluzionato il settore artistico, sollevando problematiche legate al diritto d’autore e alla speculazione. Risultati: L'uso di tecnologie digitali ha aumentato l'indipendenza degli autori dai tradizionali intermediari, con i social network come vetrine virtuali. Nonostante le incertezze giuridiche, le prospettive sono positive grazie a strumenti come smart contract e blockchain. Conclusioni: Nonostante le problematiche legate agli NFT e alle incertezze normative, l'innovazione tecnologica, come l'automazione tramite blockchain, offre opportunità per un futuro promettente per la protezione dei diritti d’autore.
The cryptocurrency market is currently one of the most interesting areas for investment, attracting both experienced and casual investors. Although it can offer high returns, it also poses significant risks due to its high volatility. In this context, artificial intelligence, particularly through deep learning and machine learning algorithms, has played a key role in developing applications that provide investment advice, with the aim of maximizing returns and reducing investment risks. This study proposes a system for forecasting the closing prices of ten of the leading cryptocurrencies currently available in the market, presented in a web application capable of making predictions ranging from one to four hours. To achieve this, different models using various machine learning and deep learning algorithms were analyzed and tested, including Recurrent Neural Networks, time series analysis algorithms such as ARIMA, and even some more conventional regression algorithms. For algorithm comparison, minute step Bitcoin price data over a 30-day period was used to forecast prices 60 minutes ahead. Through extensive experimentation, the GRU neural network demonstrated superior predictive accuracy, achieving MAPE = 0.09\%, MSE = 5954.89, RMSE = 77.17, and MAE = 60.20. A web application was also developed, which integrates the best-performing model to provide real-time price predictions for multiple cryptocurrencies.