With the expansion of cross-border e-commerce market, the demand for cosmetics anti-counterfeiting has become increasingly prominent. The traditional traceability technology is difficult to deal with the complex and changeable means of counterfeiting in cross-border e-commerce due to the problems of easy data tampering, isolated information island, and single anti-counterfeiting means. The traceability system proposed by the research takes a dual-chain consortium blockchain (a business data chain and an audit chain) as the core carrier the blockchain distributed ledger as the core carrier, and combines smart contracts to achieve automatic collection of cosmetics’ whole life cycle data and tamper proof storage, covering key links such as raw material procurement, production and processing, cross-border logistics, and terminal sales. The traceability architecture can integrate the data storage capability of blockchain with AI’s intelligent analysis capability to form a three-dimensional anti-counterfeiting system with complementary technologies. The research results can not only improve the level of cosmetics anti-counterfeiting and reduce consumer purchase risk, but also serve as a practical paradigm for the applications of blockchain and AI in cross-border e-commerce.
Cost-sensitive feature selection promises cheaper diagnosis by trading accuracy for savings in test-acquisition cost, a natural instrument for value-based procurement in public healthcare. On real clinical data, however, such promises are easily inflated by data leakage and dataset artifacts, rarely tested for in the literature. This paper therefore asks when cost-sensitive, and specifically fuzzy, feature selection actually helps. On a cohort of 1929 older adults with per-test tariffs, three selection strategies combine Extreme Gradient Boosting (XGBoost) with a fully operationalized fuzzy layer, comprising explicit membership functions, an α -cut acceptability rule, triangular fuzzy costs and Mamdani merit inference, all evaluated inside a leakage-free nested cross-validation with standard baselines and statistical tests. The central methodological contribution is a pair of reusable guards, one for target circularity and determinism, one for monotonicity and discretization artifacts. Across ten targets in eleven pool configurations, the guards show that the “about seven-fold cheaper at ≈96% accuracy” frailty result is an artifact of target circularity, that age-related hearing loss is a discretization artifact, and that most targets carry little signal in independent tests. Only dementia and central auditory processing disorder survive both guards; for dementia, a near-free education-and-age triage retains almost all the predictability of a €413 battery. Over 1000 tariff perturbations the fuzzy layer does not reorder subsets but delimits the 31.4% of comparisons that ± 15% tariff uncertainty leaves undecidable, a band containing every observed inversion. It thus contributes graded, tariff-aware decisions rather than accuracy: savings are real only where the signal survives both guards.
Brain–Computer Interfaces (BCIs) are rapidly transitioning from experimental prototypes to clinical and consumer deployments, exposing a novel attack surface: neural signal hijacking, wherein adversaries intercept, replay, or inject synthetic electroencephalographic (EEG) streams to compromise prosthetic actuation or infer private cognitive states. Conventional transport-layer security mechanisms fail to address the stochastic, non-stationary, and session-variant characteristics inherent to neural data. To bridge this gap, we introduce the Cognitive Cryptography Framework (CCF), an end-to-end security architecture that co-designs neurophysiological signal processing with cryptographic primitives for context-aware authenticated encryption. At its core, CCF employs the Oscillation-Bound Frequency-Shift Keyed Hash (OB-FSKH) algorithm, which cryptographically binds epoch-specific session keys to a subject live neural oscillation fingerprint across five canonical frequency bands ( δ , θ , α , β , γ ), ensuring automatic key expiration upon cognitive state transitions. Cryptographic entropy is harvested via a Cognitive Biometric Entropy Pool (CBEP) that distills high-fidelity min-entropy from inter-channel phase-locking values and band-power dynamics, seeding a forward-secret HKDF-SHA3-512 double-ratchet that formally guarantees IND-CCA2 confidentiality and EUF-CMA integrity under the random-oracle model. Concurrently, a lightweight Neural Anomaly Detector (NAD) leverages adaptive Mahalanobis distance scoring and Ljung–Box autocorrelation analysis to identify replay and generative injection attacks within a single 250 ms epoch. Extensive empirical validation across two benchmark datasets (BCI Competition IV Dataset 2a and PhysioNet Motor Movement/Imagery) under four adversarial scenarios demonstrates a mean authentication accuracy of 98.7
Musk Ox Optimizer (MO) is a metaheuristic based upon the social behavior of musk ox populations, which depend upon their population-wide defensive, feeding and migrating decisions based on signals provided by the "guards" within the population. It has found application use in several scientific and engineering applications with success. The present study will enhance the effectiveness of the classic MO algorithm. Its ability to innovate and exploit potentially new solutions is also aimed at enhancing its performance. This work suggests an improved MO algorithm to strike the balance between exploitation and exploration potential to attain an improved performance of traditional MO. The exploration capabilities of MO have been upgraded by the use of the Elite Opposition Based Learning (EOBL) approach. Local Neighborhood Search (LNS) has also enhanced its exploitation capabilities, inspired by the experience of the best solution observed in a local neighborhood of the current solution. The suggested approach is tested with the CEC2019 benchmark functions, and the computational results indicate that it provides promising results. Finally, to optimize the allocation of frequency spectrum, the proposed algorithm has been applied on the cognitive radio (CR) system to indicate its potential in solving real-life situations. It has been demonstrated through experimentation and statistical analysis that the suggested enhanced algorithm is a superior alternative for optimization research.
Early detection, identification and interpretation of cotton leaf disease are critical to reduce loss of the cotton crop, and can assist in decision-making for farming systems. In this regard, there exist several deep learning techniques that have been able to classify plant diseases with success, but most of the available methods have been based primarily on global image feature. To more fully capture these subtle disease characteristics, an global-local adaptive approach to cotton leaf disease classification is presented. The proposed framework uses two-branch architecture: One global contextual feature extractor and one local extractor for lesions. The global branch is utilized in capturing the context information from the entire leaf image, whereas the local branch is utilized in capturing the details of the lesions. A lesion-confidence adaptive fusion mechanism is added to adaptively adjust the contribution of features based on the lesion prominence and spatial distribution of the disease. In addition, the framework is complemented with augmentation based on lesions, where training is concentrated on diseases-relevant region. Unlike conventional static fusion approaches, the proposed adaptive mechanism performs sample-dependent feature weighting, enabling improved discrimination of morphologically similar disease categories. The key parameters are optimized with genetic algorithms and Grad-CAM combined with a consistency metric are used to enhance interpretability and obtain more accurate visual explanations of model decisions. The performance was measured with the dataset SAR-CLD-2024 in a classification context of seven classes. Experimental evaluation on the SAR-CLD-2024 dataset demonstrates that the proposed framework achieves 97.0
