To managing identities in a secure and decentralized manner, new opportunities have emerged because of recent breakthroughs in blockchain technology and biometric authentication. Blockchain is different from traditional biometric systems in that it is an unchangeable, distributed ledger that runs safe, decentralized code. Traditional biometric systems store data in one location and canât be updated. Traditional biometric systems have some flaws, including template tampering, channel interception, and comparator overrides. So, the proposed work presents a Distributed Multimodal Biometric Security System with Blockchain to handle such issues. This system uses 3D face and 3D ear biometrics with blockchain technology, which comprises IPFS, smart contracts, and decentralized applications. Features from 3D face and 3D ear are embedded into a single multimodal template, which then undergoes encryption and storage on IPFS via content-addressed storage. The Content Identifier (CID) and data are then archived by smart contracts on the blockchain to maintain data integrity, security, verifiability, and immutability. In this way, a person can prove his identity without using any central services, further improving privacy. Blockchain consensus and the smart-contract-based access control mechanism further provide security, audibility, and simplicity to P2P transactions in biometric enrolment testing results show that feature extraction takes from 120 ms to 300 ms, uploading to IPFS takes between 200 and 600 ms, and completing blockchain transactions on local private network takes from 0.5 to 1 s, using 117,519 gas per enrolment. Additional analysis on the Ethereum Sepolia test network reveals that transaction fees change depending on network conditions, but gas consumption stays deterministic. The suggested solution is resistant to typical attacks like replay, interception, and template alteration; it is also irreversible, revocable, and unlinkable, according to security analysis conducted under a formal adversarial model.
In this paper we show that the generalization error of AdaBoost is $Î\big(\tfrac{d\ln(nÎł^{2}/d)}{nÎł^2}+\tfrac{\ln(1/δ)}{n}\big)$, where $Îł$ is the advantage guaranteed by the weak learner, $d$ is the VC-dimension of the class containing the weak hypotheses, $n$ is the sample size, and $δ$ is the confidence parameter. The contribution of this paper is the upper bound; the matching lower bound follows from prior work. The upper bound proof follows by combining the known fact that AdaBoost outputs a voting classifier whose voting function has zero empirical $Îł/2$-margin loss with what is, to the best of our knowledge, a new margin-based generalization bound for voting classifiers.
Giovanni Seraghiti, KĂŠvin Dubrulle, Arnaud Vandaele, Nicolas Gillis
Nonnegative matrix factorization (NMF) approximates a nonnegative matrix, $X$, by the product of two nonnegative factors, $WH$, where $W$ has $r$ columns and $H$ has $r$ rows. In this paper, we consider NMF using the component-wise L1 norm as the error measure (L1-NMF), which is suited for data corrupted by heavy-tailed noise, such as Laplace noise or salt and pepper noise, or in the presence of outliers. Our first contribution is an NP-hardness proof for L1-NMF, even when $r=1$, in contrast to the standard NMF that uses least squares. Our second contribution is to show that L1-NMF strongly enforces sparsity in the factors for sparse input matrices, thereby favoring interpretability. However, if the data is affected by false zeros, too sparse solutions might degrade the model. Our third contribution is a new, more general, L1-NMF model for sparse data, dubbed weighted L1-NMF (wL1-NMF), where the sparsity of the factorization is controlled by adding a penalization parameter to the entries of $WH$ associated with zeros in the data. The fourth contribution is a new coordinate descent (CD) approach for wL1-NMF, denoted as sparse CD (sCD), where each subproblem is solved by a weighted median algorithm. To the best of our knowledge, sCD is the first algorithm for L1-NMF whose complexity scales with the number of nonzero entries in the data, making it efficient in handling large-scale, sparse data. We perform extensive numerical experiments on synthetic and real-world data to show the effectiveness of our new proposed model (wL1-NMF) and algorithm (sCD).
The rapid adoption of smart-home and Internet-of-Things (IoT) devices has intensified the need for privacy-preserving biometric authentication that is both secure and computationally efficient. This paper presents Hybrid-HE LLE, a practical framework that combines Locally Linear Embedding (LLE) with selective homomorphic encryption to protect face-recognition features in resource-constrained IoT environments. Unlike cloud-centric outsourcing, the proposed system performs all heavy linear-algebra operations within a semi-trusted Insider Hub, ensuring data sovereignty, low latency, and verifiable computation without revealing raw facial features. A sparse orthogonal or Toeplitz transform first obfuscates feature vectors, after which sensitive coefficients are selectively encrypted using CKKS-based polynomial encoding. Homomorphic hashing and optional zero-knowledge proofs guarantee the integrity and auditability of outsourced results. Experiments on the ORL and LFW datasets demonstrate over 94 % Rank-1 accuracy, while reducing client computation by 92 %, uplink bandwidth by 80 %, and energy usage by 55 %, with authentication latency below 120 ms on a Raspberry Pi 4-class edge device. The framework provides formal protection against IND-CPA, EUF-CMA, and IND-CCA adversaries and maintains compliance with GDPR/HIPAA requirements. Hybrid-HE LLE thus offers a scalable, secure, and real-time solution for privacy-preserving biometric access in modern IoT communication systems.
This study employs clustering analysis to evaluate the efficiency of GPUs used in cryptocurrency mining, categorizing them into distinct groups based on computational output and power consumption. Using K-Means clustering, GPUs were grouped into three clusters: low-efficiency, moderate-efficiency, and high-efficiency. High-efficiency GPUs demonstrated superior hash rates (e.g., 104.79 Mh/s for AbelHash and 218.35 Mh/s for Autolykos2) despite higher power consumption, making them ideal for high-performance mining operations. Conversely, low-efficiency GPUs exhibited lower computational output and modest energy use, highlighting opportunities for hardware upgrades or repurposing. Visualization techniques, including scatter plots and pair plots, provided clear distinctions between clusters, while a silhouette score of 0.35 indicated moderate cluster separation, suggesting areas for further refinement. The findings offer actionable insights for optimizing hardware selection, reducing operational costs, and improving energy efficiency in mining operations. Additionally, this study underscores the importance of sustainability in cryptocurrency mining and provides a foundation for future research, including the integration of additional performance metrics, exploration of alternative clustering algorithms, and development of energy-efficient mining practices. These insights contribute to the broader goal of fostering a more sustainable and data-driven approach to cryptocurrency mining.
Biometric authentication is adopted in many access control scenarios in recent years. It is very convenient and secure since it compares the userâs own biometrics with those stored in the database to confirm their identification. Since then, with the vigorous development of machine learning, the performance and accuracy of biometric authentication have been greatly improved. Face recognition technology combined with convolutional neural network (CNN) is extremely efficient and has become the mainstream of access control systems (ACSs). However, identity information and access logs stored in traditional databases can be tampered by malicious insiders. Therefore, we propose a face recognition ACS that is resistant to data forgery. In this paper, a deep convolutional network is utilized to learn Euclidean embedding (based on FaceNet) of each image and achieve face recognition and verification. Quorum, which is built on the Ethereum blockchain, is used to store facial feature vectors and login information. Smart contracts are made to automatically put data into blocks on the chain. One is used to store feature vectors, and the other to record the arrival and departure times of employees. By combining these cuttingâedge technologies, an intelligent and immutable ACS that can withstand distributed denialâofâservice (DDoS) and other internal and external attacks is created. Finally, an experiment is conducted to assess the effectiveness of the proposed system to demonstrate its practicality.
Bitcoin is a kind of Cryptocurrency and now is one of type of investment on the stock market.Stock markets are influenced by many risks of factor.And bitcoin is one kind of cryptocurrency that keep rising in recent few years, and sometimes sudden fall without knowing influence behind it on the stock market.Because it's fluctuations, there's a need and automation tool to predict bitcoin on the stock market.This research study learns how to create model prediction bitcoin stock market prediction using LSTM, LSTM (Long Short Term Memory) is another type of module provided for RNN later developed and popularized by many researchers, like RNN, the LSTM also consists of modules with recurrent consistency.The contribution of this study is providing a new forecasting framework for bitcoin price prediction can overcome and improve the problem of input variables selection in LSTM without strict assumptions of data assumption.The results revealed its possible applicability in various cryptocurrencies prediction, industry instances such as medical data or financial timeseries data.The Method that we apply on this research, also technique and tools to predict Bitcoin on stock market yahoo finance can predict the result above $ 12600 USD for next days after prediction, in the last section we make conclusions and discuss future works.The proposed methodology is then applied to train a simple Long Short Term Memory (LSTM) model to predict the bitcoin price for the upcoming 5 days.When the LSTM model is trained with a suitable data chunk, thus identified, sustainable results are found for the prediction.In the end of this paper, the work culminates with future improvements.