Journal of Computational Science最新文献

Complex systems intricately intertwine with life, and the identification of the most influential spreaders in complex networks can aid in resolving numerous pragmatic problems. Nevertheless, the identification of such kinds of nodes currently stands as an open and challenging issue. In order to accurately and efficiently address this issue, numerous metrics have been proposed. In this paper, we propose a new method based on degree, clustering coefficient and k-shell decomposition value—$DCK$ to detect the most influential spreaders by gauging the spreading ability of nodes. The proposed centrality assesses the significance of a node by the impacts of its neighbors, encompassing both the local and global network structures. To evaluate the performance of $DCK$, we compare it with different centrality measures under utilizing the Susceptible–Infected–Recovered model to simulate the propagation of epidemics across real-world networks. Experiments on real networks illustrate that $DCK$ exhibits superior differentiation ability and more accurate identification ability for influential spreaders and compared with other methods, Kendall’s $\tau$ correlation coefficient of the $DCK$ could be enhanced by 12.82%, 13.20%, 8.62%, 5.32%, 7.97% and 11.73% for the degree centrality, K-shell decomposition, $GLI$ centrality, $H$-$GSM$ centrality, $LGI$ centrality and $NPCC$ centrality.

This paper presents a numerical approach to solve third and fourth order intego-differential equations (IDEs). In order to ascertain the numerical solution for third and fourth order IDEs of second kind, the newly introduced Higher order Haar wavelet method (HOHWM) has been employed to improve the numerical result and rate of convergence compared to classical Haar wavelet approach. Some examples available in the literature have been solved to verify the HOHWM’s effectiveness. To ensure that the approach presented is legitimate, applicable and achieves its objective, the maximum absolute error of each test problem is calculated at a test point.

Nonlinear initial / boundary value problems present challenges in solving due to the divergence of coefficients near singular points. This study introduces a novel hypergeometric wavelet-based approach designed to effectively address these equations. The specialized wavelet method efficiently manages singularities, resulting in improved accuracy. To evaluate the precision and effectiveness of this approach, Lane-Emden type problems are solved using the proposed methodology and compared against established benchmarks. Comparative analyses with alternative wavelet methods are conducted, featuring absolute error tables and graphical representations. The findings highlight the exceptional accuracy and efficiency of the proposed method relative to existing approaches. An advantage of this method is its requirement of fewer basis functions, leading to reduced computational time and complexity.

Regression of potential energy functions stands as one of the most prevalent applications of machine learning in the realm of materials simulation, offering the prospect of accelerating simulations by several orders of magnitude. Recently, graph-based architectures have emerged as particularly adept for modeling molecular systems. However, the development of robust and transferable potentials, leading to stable simulations for different sizes and physical conditions, remains an ongoing area of investigation. In this study, we compare the performance of several graph neural networks for predicting the energy of water cluster anions, a system of fundamental interest in Chemistry and Biology. Following the identification of the graph attention network as the optimal aggregation procedure for this task, we obtained an efficient and accurate energy model. This model is then employed to conduct Monte Carlo simulations of clusters across different sizes, demonstrating stable behavior. Notably, the predicted surface-to-interior state transition point and the bulk energy of the system are consistent with findings from other investigations, at a computational cost three-orders of magnitude lower.

The space fractional Klein–Gordon-Zakharov equations are transformed into the multi-symplectic structure system by introducing new auxiliary variables. The multi-symplectic system, which satisfies the multi-symplectic conservation, local energy and momentum conservation, is discretizated into the semi-discrete multi-symplectic system by the Fourier pseudo-spectral method. The second order multi-symplectic average vector field method is applied to the semi-discrete system. The fully discrete energy preserving scheme of the space fractional Klein–Gordon-Zakharov equation is obtained. Based on the composition method, a fourth order energy preserving scheme of the Riesz space fractional Klein–Gordon-Zakharov equations is also obtained. Numerical experiments confirm that these new schemes can have computing ability for a long time and can well preserve the discrete energy conservation property of the equations.

Background and Objective:

The constant growth of invasions and information theft by using infected software has always been a problem. According to McAfee labs in 2020, on average, 480 new viruses are created each hour. The means of identifying such threats, categorizing and creating vaccines may not be that fast. Thanks to the increasing processing power and the popularity of artificial intelligence, it is now possible to integrate intelligence on an antivirus engine to enhance its protecting capabilities. And doing so with good algorithms and parameterization can be a key asset in securing one’s environment. In this work we analyze the overall performance of our antivirus and compare it with other state-of-art antiviruses.

Methods:

In this work, we create an extreme neural network which can perform quick training time and have satisfactory accuracy when classifying unknown files that may or may not be infected with Citadel. Our virus database is built with many examples of well-known infected files, and our results are compared with other intelligent antiviruses created by other companies and/or researchers.

The proposed technique stands out as a beneficial practice in terms of efficiency and interpretability; it achieves a very reduced number of neurons through its thorough pruning process. This reduction of dimensionality shrinks the input layer by 98%, enhancing not only data interpretation but also reducing the time required for training.

Results:

Our antivirus achieves an overall performance of 98.50% when distinguishing harmless and malicious portable executable (PE) programs. To enhance accuracy, we conducted tests under various initial conditions, learning functions, and architectures. Our successful results consumes only 0.19 s of training when using the complete training database and the response time is so immediate that the computer rounds it to 0.00 s.

Conclusions:

In this work, we conclude that mELM implementations are viable, and their performance can match state-of-the-art ones. It’s training and classification times are among the fastest of the algorithms tested, and the accuracy in detecting Citadel-infected PEs is acceptable.

In this paper, we introduce a novel hybrid approach that leverages both data and numerical simulations to address the challenges of solving forward and inverse wave problems, particularly in the mid-frequency range. Our method is tailored for efficiency and accuracy, considering the computationally intensive nature of these problems, which arise from the need for refined mesh grids and a high number of degrees of freedom. Our approach unfolds in multiple stages, each targeting a specific frequency range. Initially, we decompose the wave field into a grid of finely resolved points, designed to capture the intricate details at various wavenumbers within the frequency range of interest. Importantly, the distribution of these grid points remains consistent across different wavenumbers. Subsequently, we generate a substantial dataset comprising 1,000 maps covering the entire frequency range. Creating such a dataset, especially at higher frequencies, can pose a significant computational challenge. To tackle this, we employ a highly efficient enrichment-based finite element method, ensuring the dataset’s creation is computationally manageable. The dataset which encompasses 1000 different values of the wavenumbers with their corresponding wave simulation will be the basis to train a fully connected neural network. In the forward problem the neural network is trained such that a wave pattern is predicted for each value of the wavenumber. To address the inverse problem while upholding stability, we introduce latent variables to reduce the number of physical parameters. Our trained deep network undergoes rigorous testing for both forward and inverse problems, enabling a direct comparison between predicted solutions and their original counterparts. Once the network is trained, it becomes a powerful tool for accurately solving wave problems in a fraction of the CPU time required by alternative methods. Notably, our approach is supervised, as it relies on a dataset generated through the enriched finite element method, and hyperparameter tuning is carried out for both the forward and inverse networks.

Insulin is a kind of protein that regulates the blood sugar levels is significant to prevent complications associated with diabetes, such as cancer, neurodegenerative disorders, cardiovascular disease, and kidney damage. Insulin protein (IP) plays an active role in drug discovery, medicine, and therapeutic methods. Unlike experimental protocols, computational predictors are fast and can predict IP accurately. This work introduces a model, called IP-GCN for IP prediction. The patterns from IP are extracted by K-spaced position specific scoring matrix (KS-PSSM) and the model training is accomplished using powerful deep learning tool, called Graph Convolutional Network (GCN). Additionally, we implemented Pseudo Amino Acid Composition (PseAAC) and Dipeptide Composition (DPC) for feature encoding to assess the predictive performance of GCN. To evaluate the efficacy of our novel approach, we compare its performance with well-known deep/machine learning algorithms such as Convolutional Neural Network (CNN), Extremely Randomized Tree (ERT), and Support Vector Machine (SVM). Predictive results demonstrate that the proposed predictor (IP-GCN) secured the best performance on both training and testing datasets. The novel computational would be fruitful in diabetes drug discovery and contributes to research for therapeutic interventions in various Insulin protein associated diseases.

In the present scenario, automatic Human Activity Recognition (HAR) is an emerging research topic, particularly in the applications of healthcare, Human Computer Interaction (HCI), and smart homes. By reviewing existing literature, the majority of the HAR methods achieved limited performance, while trained and tested utilizing unseen Internet of Things (IoT) data. In order to achieve higher recognition performance in the context of HAR, a new clustering method named Modified Differential Evolution based Fuzzy Clustering (MDEFC) is proposed in this article. The proposed MDEFC method incorporates an asymptotic termination rule and a new differential weight for enhancing the termination condition and improving this method’s ability in exploring the solution space of the objective function. The extensive empirical analysis states that the proposed MDEFC method achieved impressive recognition results with minimal training time by using both spatial and temporal features of the individual. The proposed MDEFC method’s effectiveness is tested on a real time dataset and an online Wireless Sensor Data Mining (WISDM) v1.1 dataset. The result findings demonstrate that the proposed MDEFC method averagely obtained 99.73 % of precision and 99.86 % of recall on the WISDM v1.1 dataset. Similarly, the proposed MDEFC method averagely obtained 93.46 % of f1-measure, 94.60 % of recall, and 93.88 % of precision on the real time dataset. These obtained experimental results are significantly higher in comparison to the traditional HAR methods.

This study analyses the sensitivity analysis of the friction factor and heat transfer rate within a hybrid nanoliquid flow of 20W40 motor oil (a base liquid that has been characterized by the Society of Automotive Engineers) + nickel zinc ferrite- manganese zinc ferrite over a stretchable sheet utilizing the Response Surface Methodology (RSM) along with irreversibility analysis. The melting phenomenon with buoyancy effect has been considered. Hybrid nanofluids exhibit improved thermal connectivity, enhanced mechanical resilience, favorable aspect ratios, and superior thermal conductivity when compared to conventional nanofluids. The system of governing equations is transformed into dimensionless form using the Lie group approach. Numerical computations are performed utilizing the spectral local linearization method. It is demonstrated that the Nusselt number and friction drag are decreased due to the increase of manganese and nickel zinc ferrites particles in the fluid. Further, the melting parameter reduces entropy generation by 41.16% and the viscous dissipation parameter minimizes surface friction. Sensitivity analysis, conducted through RSM, reveals that skin friction and the Nusselt number are positively sensitive to the melting parameter. The numerical solutions have been compared with the available results along with error estimations, which show excellent agreement. Comparison of both hybrid nanofluids are displayed graphically. Finally, this work has many uses such as microwave and biomedical applications, electromagnetic interfaces, melting, and welding operations which are the most significant manufacturing applications important in various sectors such as cooling systems of nuclear reactors.