Plagiarism is prevalent in various fields, particularly in academia and research, which makes its detection challenging. Deep Learning (DL) methods have great potential for developing the accuracy of similarity detection in text documents. However, recent studies indicate a lack of extensive research utilizing large datasets and human input. The current research is largely based on natural language processing to predict the similarity of the text between original and suspected plagiarized text, which leads to better prediction results. The proposed Weight-Optimized Dense Neural Network–Bidirectional Long Short-Term Memory system combines optimized dense representations with bidirectional contextual learning to detect plagiarism. The Slender Loris Optimization Algorithm is chosen to optimize the model parameters since it favors good parameters exploration and exploitation within the weight space and the selection of discriminative model configurations. The proposed hybrid DL system is implemented in Python with the TensorFlow library, and the accuracy, precision, recall, and F1 score are compared with previous text similarity detection algorithms. The results prove that the proposed system outperforms the existing ones with terms of text similarity identification with an accuracy of 97.03
Picture fuzzy multigroups were recently proposed by combining picture fuzzy multisets with fuzzy group theory. We develop a modified framework rather than a new antecedent notion. The modification addresses scalar inequalities applied to sequence valued memberships and independent optimization of the three picture fuzzy coordinates. We prove that marginal coordinate sequences do not determine admissible occurrence triples and that universal preservation of the picture fuzzy simplex under reindexing forces one common permutation of the positive, neutral, and negative coordinates. This provides a mathematical justification for occurrence alignment. We then define fixed height occurrence aligned picture fuzzy multigroups and distinguish inherited levelwise subgroup results from new cross level results. In the order coherent subclass, occurrence cuts form nested subgroup filtrations. Their survival counts are bounded multigroups, and the complete family of these threshold multigroups reconstructs the original picture fuzzy object. The threshold construction is compatible with meets, direct products, pullbacks, and quotients. We also establish a closure counterexample for independent convolution, cut and normal cut characterizations, arbitrary meet closure with possible failure of joins, and a global core/saturation quotient correspondence. Finite verification requires O(m|G|^2) time. A height three example on the nonabelian group D_4 yields multiplicities 1, 2, 3, nested normal cuts, and a quotient by ⟨ r^2⟩ .
We study logarithmic power series summability of double series. We first prove an Abelian theorem showing that bounded convergence of a double series implies summability to the same value by the logarithmic power series method. We then establish a Tauberian converse assuming two coordinatewise first-difference conditions of logarithmic order. Examples are given to show that logarithmic power series summability alone does not imply convergence and that both coordinatewise Tauberian conditions are essential. Finally, we apply the Tauberian theorem to double Fourier series and obtain conditions on the Fourier coefficients under which logarithmic power series summability implies convergence at a fixed point.
In this study, a nonlinear fractional gas dynamics model is inspected using both analytical and numerical measures. Exact travelling-wave solutions of such equation is realized using the Generalized Kudryashov Method (GKM) and a power series solution is assessed and linked with the GKM solution to dichotomize the accuracy and convergence of the analytical outcomes. The stability behaviour of the system is indended by checking the perturbations around the equilibrium point, leading to an explicit expression for the eigenvalue governing the stability landscapes. A bifurcation analysis is carried out to formulate the critical parameter settings under which qualitative instabilities in the system dynamics arise. Sensitivity analysis with respect to the parameters k, , and α is also accomplished to review their impression on the solution behaviour. Numerical simulations are exposed to find the analytical conclusions and to picturize the travelling-wave structures of the gas dynamics model. Graphical comparisons between the GKM and the power-series solutions designate excellent closedness, evidencingthe unassailability of the proposed analytical progression. The results admit that the proposed analytical model yields valuable insights into the nonlinear wave propagation and stability landscapes of fractional gas dynamics classifications.
The hydraulic performance of submerged hydraulic jumps is significantly altered with provision of channel bed macroroughness. Earlier studies primarily considered idealized shape roughness, whereas non-idealized shape roughness that better represents field requirements have received limited attention. The present study investigated six macroroughness configurations considering idealized as well as non-idealized shape of roughness for an inflow Froude number ( Fr_1 ) range of 3.5–8.1 and proposes a dimensionless macroroughness shape parameter ( Ψ ), combining geometric compactness and cavity-induced recirculation potential to represent strip-type macroroughness. A two-dimensional Computational Fluid Dynamics (CFD) model was developed in ANSYS Fluent using the Volume of Fluid (VOF) method coupled with Renormalization Group (RNG) k-ϵ turbulence model. The model was validated against laboratory measurements of longitudinal velocity, tailwater depth ratio and relative roller length. The CFD model showed good agreement with experimental results, with coefficient of determination ( R^2 ) = 0.876–0.993 for velocity profiles, R^2 = 0.996–0.998 for tailwater depth ratio and relative roller length. The tailwater depth ratio and relative roller length showed mean absolute percentage error (MAPE) values of 1.72 Ψ ) varied from 2.86 to 10.54, corresponding to the investigated range of geometric complexity and cavity-induced recirculation potential. The findings indicated an apparent transition around Ψ ≈ 5.71 , beyond which the reductions in tailwater depth ratio and relative roller length became marginal within the investigated range. These findings demonstrate that Ψ provides a practical, physically interpretable geometric descriptor for evaluating, comparing idealized and non-idealized macroroughness configurations, thereby offering a useful basis for hydraulic design and optimization. The proposed parameter may also facilitate the development of generalized predictive models for submerged hydraulic jumps within the investigated range of strip-type macroroughness configurations.
Extreme weather events such as floods have a significant impact on society and livelihood because of their devastating nature, causing loss of life, damage to agriculture, and adverse effects on health. Although various studies have been conducted on flood mitigation strategies, research in the context of climate change has been significantly lacking. The present study aims to assess the severity and temporal trends of meteorological flood events over the Indian region under two climate change scenarios: shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5). The flood characteristics were studied using precipitation data from the Coupled Model Intercomparison Project of General Circulation Models for the historical (1985–2014) and future period (near future (2030–2059) and far future (2070–2099)). In addition, IMD observational gridded precipitation data have been considered for the evaluation of historical CMIP6 precipitation and its multi-model ensemble mean (MME). The flood events over India (AI) and the core monsoon zone (CMZ) were characterized using the Standardized Precipitation Index (SPI). Statistical analyses suggest that, despite its systematic dry bias, the MME performs better compared to the individual CMIP6 models. Further analysis of MME precipitation revealed that the severity of wet events is notably rising in the SSP5-8.5 scenarios compared with SSP2-4.5. Furthermore, extreme wet events exhibit an increasing trend in the FF compared to the NF under both emission scenarios over India. In contrast, severe wet events show a decreasing trend over the CMZ, particularly in the FF under the SSP5-8.5 scenario. This study has important implications on agriculture activity, water resource management, and flood mitigation strategies.
Fractal–fractional differential equations provide an effective mathematical framework for modeling complex dynamical systems exhibiting both memory effects and fractal characteristics. In this work, a novel iterative method is developed for solving fractal–fractional differential equations with exponential and Mittag–Leffler memory kernels by extending the fractional Daftardar–Gejji and Jafari New Iterative Method. The proposed approach provides a simple and computationally efficient framework for obtaining approximate solutions to both linear and nonlinear problems without requiring discretization, perturbation techniques, linearization, or auxiliary parameters. A theoretical convergence and stability analysis is established under suitable assumptions using the Lipschitz condition and the boundedness of the fractal–fractional integral operator, ensuring the convergence of the iterative sequence to the unique solution. The effectiveness of the proposed method is demonstrated through several benchmark fractal–fractional models, including diffusion, nonlinear diffusion, Korteweg–de Vries, and Navier–Stokes-type equations with exponential and Mittag–Leffler memory kernels. Numerical simulations, graphical comparisons, and convergence analysis confirm that the proposed method yields accurate and stable approximate solutions while effectively capturing the influence of the fractional order and fractal dimension on the solution behaviour. The obtained results demonstrate that the proposed iterative method is reliable, computationally efficient, and broadly applicable for solving a wide class of fractal–fractional differential equations arising in applied mathematics, physics, and engineering.
In this paper, we analyze a non-Markovian queueing model providing service in up to K sequential phases. After completing each service phase, a customer either proceeds to the next with probability q_i or exits the system with probability q_i=(1-q_i) . The model incorporates the possibility of server breakdowns, where the server may fail randomly and is then sent for repair. Customers arrive in batches of random size according to a Poisson process. If the server is busy upon arrival, they join a retrial group and attempt re-entry after a random time. Customers are impatient and may balk if the server is busy or under repair. Additionally, once the system becomes empty, the server takes a vacation; if still idle upon return, it may take one additional optional vacation selected from among m available options. The Supplementary Variable Technique (SVT) is employed to derive system performance measures, and numerical experiments are conducted to perform sensitivity analysis.
Sensitive quantitative characteristics, such as household income, incidence of premarital abortion or substance use, are difficult to reliably measure, because respondents frequently distort or withhold their true value for fear of stigma. Scrambled response (SR) techniques address the issue by allowing a respondent to mask an answer by a randomization device before it is reported. However, current SR estimators for heterogeneous populations have been mostly developed either under simple random sampling or under unstratified ranked set sampling and have generally used a single additive or multiplicative scrambling mechanism rather than a combined one. In this paper, we develop a new scrambled response strategy under stratified ranked set sampling (SRSS). The sensitive response Y is scrambled through a combined multiplicative-additive transformation Z = UY + V, which includes the classical additive [21] and multiplicative [11] scrambling schemes as special cases. The scrambled output data are used to propose a class of almost unbiased ratio type estimators of the population mean indexed by three scalars chosen so that the weights sum to one, the first order bias vanishes and the mean square error is minimized; the bias and mean square error of the resulting estimator are derived to the first order of approximation and the optimum member of the class is obtained in closed form. The efficiency of the proposed estimator is assessed in terms of percent relative efficiency and percent relative loss with respect to the corresponding non-scrambled stratified ranked set sampling estimator based on a real population of U.S. state-level abortion rates and an artificially generated population under several configurations of the scrambling-variable parameters. In all the configurations considered, the proposed estimator is more efficient than the non-scrambled one, which means that the loss of precision incurred by scrambling to protect the respondent’s privacy is more than compensated by the joint use of stratification, ranking and the flexible three-parameter estimator class. The results provide a basis for recommending the proposed strategy for surveying practitioners who deal with sensitive quantitative data from heterogeneous populations. The paper ends with a discussion of extensions to imperfect ranking and higher-order approximations as future work.
Breast cancer remains a leading cause of mortality among women worldwide, underscoring the need for accurate and early diagnosis. Although mammography and ultrasound are widely used in clinical practice, most Computer-Aided Diagnosis systems rely on single-modality analysis, resulting in limited robustness, noise sensitivity, and reduced generalization across heterogeneous datasets. To address these limitations, this research proposes a unified multimodal network that integrates complementary information from mammography and ultrasound to enhance diagnostic reliability. The framework incorporates Region of Interest (ROI) extraction, data augmentation, modality-specific noise filtering, and deep feature extraction using a transfer learning-based dual convolutional neural network with vision transformer architecture. Feature relevance is refined through hybrid optimization and Quality-Adaptive Feature Gating (QAFG), followed by hierarchical fusion and classification using a reinforcement learning-based strategy. Experiments on three datasets demonstrate strong performance, achieving 99.45
We introduce and study the deferred logarithmic summability method for sequences of fuzzy numbers. This new method provides a broader framework for the analysis of slowly convergent or divergent fuzzy sequences under uncertainty. We first prove that the proposed method is regular and investigate its relationship with the classical Cesàro and logarithmic summability methods. Several inclusion theorems are established, showing that deferred logarithmic summability is strictly stronger than both (C, 1) and (ℓ ,1) summability in the fuzzy setting. Furthermore, we obtain necessary and sufficient Tauberian conditions (including a slow-decrease-type condition) under which deferred logarithmic summability implies classical convergence of the sequence or its subsequences. The results unify and extend several existing summability concepts in fuzzy analysis and offer new tools for applications in approximation theory, operator theory, and uncertainty modeling.
Circular distributions play a critical role in modelling data on the unit circle and are therefore well-suited for directional phenomena. In this paper, we propose the Weighted Wrapped Stable (WWS) distribution, a new family of circular distributions formed by embedding a weight function within the wrapped stable framework. The WWS model is designed to capture asymmetric and positively skewed directional data, addressing a key limitation of existing symmetric circular stable distributions. The inclusion of a weighting mechanism enhances flexibility, enabling the WWS family to model more complex structures in directional statistics. We also develop the theoretical foundation of the proposed distribution. Specifically, we derive closed-form expressions for the characteristic function and trigonometric moments, and examine several key structural properties. Parameter estimation is performed using the Maximum Likelihood Estimation (MLE) method. The performance and consistency of the WWS distribution are evaluated through Monte Carlo simulations and illustrated using two real-world datasets. Model fit and predictive accuracy are assessed based on multiple goodness-of-fit criteria. Comparative results confirm that the WWS distribution is an effective, flexible, and practically useful tool for modelling asymmetric directional data.
The Kameng watershed, located in the western Arunachal Himalaya, represents a structurally complex and tectonically active region that extends across the major Himalayan thrust belts: Main Frontal Thrust (MFT), Main Boundary Thrust (MBT), Main Central Thrust (MCT), and South Tibetan Detachment System (STDS). This study evaluates spatial variations in landscape evolution using hypsometric integrals (HI) and hypsometric curves derived from 30 m COP-DEM data for 17 sub-watersheds, supported by regional lithological, seismic, and published GPS data. HI values range from 0.477 to 0.505, broadly categorising all sub-watersheds within the mature geomorphic stage. Although differences among the five Himalayan subdivisions are not statistically significant (H = 9.150), (df = 4), (p = 0.057), the southern Brahmaputra Plain and the Siwalik sub-watersheds generally exhibit lower HI values and concave curves, whereas S-shaped curves predominantly characterise the Lesser Himalayan sub-watersheds. Greater and Tethyan Himalayan sub-watersheds commonly exhibit S-shaped to locally convex curves and comparatively high normalised elevations. The dimensionless relative uplift index (Ut) varies from 0.53 to 1.03 but is interpreted as an integrated geomorphological index rather than an actual uplift value. Collectively, these spatial patterns indicate relative differences in relief preservation and landscape adjustment arising from the combined influences of tectonics, lithology, fluvial denudation, and local glacial inheritance. Published GPS and seismic data provide qualitative regional evidence of ongoing deformation but do not quantitatively validate the geomorphic indices.
In this paper, we introduce a new Banach space (C(ℝ_+, L^p(ℝ_+)), ‖·‖ _C_ϕ) equipped with a weighted norm depending on a function ϕ , and we characterize the compact subsets of this space. Based on this characterization, we define a new measure of noncompactness adapted to this functional framework. Using this measure of noncompactness together with Darbo’s fixed point theorem, we study the solvability of a higher-order Caputo fractional differential equation on an unbounded domain with nonlocal boundary conditions. The main novelties of this work are the introduction of the weighted Banach space C(ℝ_+, L^p(ℝ_+)) , which is particularly suitable for problems on unbounded domains; the construction of a new measure of noncompactness that captures both the lack of equicontinuity and the lack of uniform decay at infinity; and the analysis of a new class of higher-order fractional boundary value problems. Finally, we provide a concrete example to illustrate the applicability of our main result.
The proposed SEIQRDP epidemic model for discrete-time cases with memory is introduced based on the Grünwald-Letnikov fractional difference scheme. Vaccinations are considered by incorporating the transition rate from susceptibles to the protected class. Mathematical validity is attained for the model by means of proving positivity and boundedness. The basic reproductive number is evaluated, and the local stability of the trivial steady state is attained using the Jury criterion for stability. The proof of transcritical bifurcation is also provided when the threshold condition for the epidemic is reached. Global stability of equilibria is also analyzed. Numerical findings confirm theoretical results and show that the inclusion of memory and more protection significantly reduces infections.
In this paper, we introduce the concept of ideally slowly oscillating sequences of complex uncertain variables, motivated by the classical theory of slowly oscillating sequences and their ideal extensions. We establish fundamental properties of these sequences and examine their convergence behavior within the framework of uncertainty theory. Several illustrative examples are provided to demonstrate the applicability of the results. The study not only broadens existing convergence theory under uncertainty, but also opens potential avenues for applications in approximation theory, decision-making models, and uncertain differential equations.
The rapid increase in deployment of Internet of Things (IoT) edge devices has greatly amplified the risk of on-chip cryptographic operations being compromised through side-channel attacks that exploit power, timing, and microarchitectural leakages. To tackle these multidomain threats, this article describes an adaptive secure RISC-V processor with a multi-layer defense system combining pipeline-level AES hardening, algorithmic randomization, clock-domain hiding, and speculative execution protection. Our design combines a single-cycle AES hardware accelerator with ISA extensions for custom instructions that not only make the encryption process shorter and faster but also help in fighting power analysis-based attacks. The adaptive AES hardware allows changing between AES-128, AES-192, and AES-256 on the fly, This way power profiles get more complex and the power analysis attacks are Much reduced, Mainly CPA attacks. Experimentally, the SNR is cut in half to 0.65 and MTD (minimum traces to disclosure) is raised to 110k. Besides, side-channel attack surface is also lessened by using spread-spectrum frequency hopping clocking system based on MMCM (Mixed Mode Clock Manager)-type dynamic reconfiguration, which gives time obfuscation with virtually no extra hardware. Speculative execution vulnerabilities are addressed by implementing SCSGuardian method which limits speculatively executed unsafe memory accesses and stops the leaked transient cache-states from Spectre- and Meltdown-type attacks. The whole system is capable of delivering an encryption data rate of 2.35 Gbps but the area overhead is only 17.8
This study reports the physical properties and the supercapacitor applications of environmentally benign spinel cobalt oxide (Co3O4) nanoparticles, prepared using microwave assisted, chemical co-precipitation method. The physical properties of Co3O4 nanoparticles such as structural, optical, elastic, electrical, thermal, magnetic and electron spin resonance (ESR) properties have been evaluated and elucidated. The prepared cobalt oxide nanoparticles material exhibits two optical band gaps 1.5 and 2.5 eV, p-type semiconductivity driven by small polaron hopping, and room-temperature paramagnetic behavior. The elastic properties estimated from characteristic Fourier-transform infrared (FTIR) spectral bands yielded a Young's modulus of 202.83 GPa and a Debye temperature of 572.20 K. The Co3O4 electrode exhibits maximum specific capacitance value of 29 F/g at the scan rate of 5 mV/s. This comparatively lower electrochemical performance is attributed to particle aggregation and a low specific surface area, which restrict the available electrochemically active sites. Overall, this work provides a fundamental link between microstructural parameters and functional behavior, facilitating the fundamental understanding for the development of magnetic semiconducting Co₃O₄ based materials for technological and supercapacitor applications.
Steganography is a technique of camouflage to hide secret data inside media files for covert communication. Among the media files, images are the most commonly used, as they are frequently transferred over the internet. This paper uses images as cover objects to present a secure and imperceptible steganography technique that uses music-inspired feature representation and adaptive Least Significant Bit (LSB) matching to conceal a secret image. The proposed technique adaptively embeds the secret image inside the cover image by utilizing cover image characteristics, including texture, entropy, and edge information, together with keyed pixel permutation to determine secure embedding locations. To enhance security, the secret image is first serialised, compressed, and encrypted with AES-256-GCM using the HMAC-based Key Derivation Function (HKDF) before embedding. The statistical features derived from the secret image are converted into a music-inspired representation for adaptive embedding, preserving the statistical characteristics of the cover image and minimizing visual distortion. The results obtained ensure high visual quality without perceptual distortion, making it an efficient technique for concealing data in covert multimedia communication applications. The experimental analysis shows a high Peak Signal-to-Noise Ratio (PSNR) value of above 47 decibels (dB) and SSIM values exceeding for all the considered images, with an average embedding capacity of 1.85 Bits Per Pixel (BPP). Furthermore, the proposed technique has been tested against different attacks, including noise, JPEG compression and geometrical attacks, to ensure robustness and security. Additionally, statistical analysis is also done using histogram analysis, Chi-square test, p-value, HCF Shift and HCF Relative Shift. Comparative analysis against state-of-the-art techniques demonstrates that the proposed framework provides an effective trade-off between embedding capacity and image quality. The code and the data associated with the experimental analysis are available at https://github.com/x1o3/v0xen.git .
The integration of renewable energy sources and Electric Vehicles (EVs) into the grid is essential for reducing energy costs and emissions. While unidirectional chargers provide basic grid support, bidirectional charging systems enable advanced Vehicle-to-Grid (V2G) functionality. However, limited research addresses gaps, including limited harmonic suppression and dynamic instability in bidirectional solar Photovoltaic (PV) connected EV systems under conventional controllers in power quality enhancement. This research introduces a novel optimized control framework that combines a Fractional Order-based Proportional Tilt Integral Derivative (FOPTID) controller, tuned via an opposition-based learning strategy with the puma optimizer and greedy selection, and a two-Degree-Of-Freedom Tilt-Integral-Derivative Controller with Fractional Derivative (2DOF-TIDμ) controller with fractional derivative for voltage source converter control. The proposed system significantly enhances power quality and converter reliability during bidirectional power flow. Simulation results under intermittent solar conditions show Total Harmonic Distortion (THD) reduced to 0.3 to 0.8
