Journal of Computational Science最新文献

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.

The coherence of the axial force between steel struts during excavation, i.e. the axial force coherence, is a critical factor affecting axial force control. This study introduces a novel method for calculating the horizontal displacement of a diaphragm wall, applicable to servo-controlled excavation, based on the non-limit earth pressure theory. Furthermore, the study investigates the coherence of axial forces in prestressed struts. First, the diaphragm wall is modeled as a rectangular thin plate with two opposite edges simply supported and the other two edges free. It is then divided into $m\times n$ small rectangles along the depth and length directions, and the external combined force within each small rectangle is calculated. Secondly, a non-linear set of force–displacement equations is constructed, and the recursive equation of the displacement of the diaphragm wall is obtained by applying the Newton–Raphson method. The method’s accuracy is confirmed through field measurement comparisons. Subsequently, The paper then applies the proposed methodology to scrutinize the effects of loading on individual and multiple struts on the axial forces of adjacent struts. The loading scheme for struts in deep excavation within soft soil areas can be referenced by utilizing this method.

Ecological modeling is an effective tool for studying the interactions between predator and prey species by considering various functional responses and ecological effects. Employing computational or mathematical models allows us to determine the impact of specific human interventions or animal behaviors on the evolution of these species. The fear experienced by prey due to the presence of predators plays a crucial role in shaping the dynamics of their interactions. The manuscript focuses on developing and examining a novel prey–predator model that considers predation fear, prey refuge, and anti-predator effects. Caputo fractional derivative is utilized in the construction and analysis of the model, which integrates ecological principles like memory effects to improve our comprehension of species relationship. The study investigates aspects such as stability, well-posedness, and solution uniqueness for the proposed model. We carried out extensive numerical simulation to support theoretical results. Graphical results are provided for the model encompassing a broad spectrum of fractional order values. The effect of the fear level, growth rate of prey, saturation rate and prey refuse on the behavior of the solution are discussed.

Quantum computing is an emergent computational technology having potential to solve complex computational problems. This revolutionary domain spurs intense exploration within the research community, necessitating a thorough examination to delineate scientific trajectories and glean research advancements. In this study, a scientometric investigation of quantum computing research limited to recent decade (2014-2023) is conducted sourced from the Scopus database. Analyses of publication patterns, top-cited articles, article co-citation, and author co-citation are performed to elucidate research trajectories. The results highlight that qubit quality control is a major focus of study, along with research in quantum cryptography, quantum neural networks, quantum annealing, and quantum-classical/classical-quantum algorithms.

To improve the performance of intelligent products and reasonably distribute the load of the loading robot, a multi-objective, and multi-objective (Traveling-salesman-problem, TSP) mathematical model was established. A genetic algorithm based on speed invariant and the elite algorithm is proposed to solve the multi-TSP assignment problem. To ensure the integration of the population, a population resettlement strategy with elite lakes was proposed to improve the probability of population transfer to the best Pareto solution. The experiment verified that this strategy can approach the optimal solution more closely during the population convergence process, and compared it with traditional Multi TSP algorithms and single function multi-objective Multi TSP algorithms. By comparing the total distance and maximum deviation of multiple robot systems, it is proven that this algorithm can effectively balance the path length of each robot in task allocation. From the research results, it can be seen that in genetic algorithms, resetting the population after reaching precocity can maintain the optimization characteristics of the population and have a high probability of obtaining Pareto solutions. At the same time, storing elite individuals from previous convergent populations for optimization can better obtain Pareto solutions.

The growth of human population and emergence of pandemics has enhanced the need for healthcare treatment and medications. The development of nanotechnology acts as a platform in diagnosis and detection of various diseases. The presence of Nano-sensors in Internet of Things (IoT) paradigm has the ability to sense and monitor real time data in various fields of applications, particularly in healthcare. In this paper, a comprehensive investigation of numerous studies that have worked on various systems for integrating Nano-sensor communication networks with the IoT in medical fields are studied. This research highlights and analyses the capabilities of various nano-forms of nano layered materials utilized in the identification of diseases. The efficiency of different techniques is validated in terms of energy consumption, detection ability and adapting ability of various environments. Moreover, this work focuses on the applications of nano-sensors communications, and networking for healthcare systems, along with challenges and topics which are needed to be explored.