Scientific Programming

Cloud computing plays a pivotal role in sharing resources and information. It is challenging to secure cloud services from different intruders. Intrusion detection system (IDS) plays a vital role in detecting intruder attacks, and it is also used to monitor the traffic in the network. The paper is aimed to control the attacks using the machine learning (ML) technique integrated with the artificial bee colony (ABC) named Group-ABC (G-ABC). The IDS detector has been implemented and further simulation results have been determined using the G-ABC. The evaluation has been carried out using the measures such as precision, recall, accuracy, and F-measure. Different attacks such as user to root (U2R), probe, root to local (R2L), backdoors, worms, and denial-of-service (DoS) attacks have been detected. The simulation analysis is performed using two datasets, namely, the NSL-KDD dataset and UNSW-NB15 dataset, and comparative analysis is performed against the existing work to prove the effectiveness of the proposed IDS. The objective of the work is to determine the intruder attacker system using the deep learning technique.

With the rapid development of Internet technology, mankind has entered the era of big data. The Internet records all kinds of information, and the amount of information generated in the future is in a state of explosive growth. With the development of enterprises, the data of accounting information have grown, and its security has also been threatened. Once the accounting information is leaked, it will have a great impact on the company. Therefore, it is very necessary to protect accounting information data. This paper aims to study how to protect accounting information data based on big data and cloud computing. Based on this, this paper proposes an attribute-based encryption method based on big data and cloud computing and analyzes the ABE scheme of a single authority and multiple authorities. The multiauthority ABE scheme can effectively solve the problems of low computational efficiency, poor security performance, and high time overhead in the single-authority ABE scheme. The experimental results of this paper show that, under the single-authority ABE scheme in the information data protection of enterprise A, the confidentiality of data is 54% at the highest and 46% at the lowest, which is generally not very high. In the information data protection of the multiauthority ABE scheme in enterprise A, the highest data confidentiality is 86% and the lowest is 79%. The highest is 32% more than the single-authority ABE scheme. The lowest is also 33% more than that. Generally speaking, the confidentiality of its data is very high. In terms of reliability, integrity, and effectiveness, the single-authority ABE scheme is not as good as the multiauthority ABE scheme. It shows that the multiauthority ABE scheme proposed in this paper has a powerful information data security protection function and can be effectively applied to accounting information data protection.

The current research offers an enhanced three-parameter lifetime model that combines the unit Burr XII distribution with a power series distribution. The novel class of distribution is named the unit Burr XII power series (UBXIIPS). This compounding technique allows for the production of flexible distributions with strong physical meanings in domains such as biology and engineering. The UBXIIPS class includes the unit Burr XII Poisson (UBXIIP) distribution, the unit Burr XII binomial distribution, the unit Burr XII geometric distribution, and the unit Burr XII negative binomial distribution. The statistical properties of the class include formulas for the density and cumulative distribution functions, and limiting behaviour, moments and incomplete moments, entropy measures, and quantile function are provided. For estimating population parameters and fuzzy reliability for the UBXIIP model, maximum likelihood and Bayesian approaches are studied by the Metropolis–Hastings algorithm. For maximum likelihood estimators, the length of asymptotic confidence intervals is specified, whereas, for Bayesian estimators, the length of credible confidence intervals is assigned. A simulation investigation of the UBXIIP model was established to evaluate the performance of suggested estimates. In addition, the UBXIIP distribution is explored using real-world data. The UBXIIP distribution appears to offer some benefits in understanding lifetime data when compared to unit Weibull, beta, Kumaraswamy, Kumaraswamy Kumaraswamy, Marshall-Olkin Kumaraswamy, and Topp–Leone Weibull Lomax distributions.

A synthetic aperture radar (SAR) automatic target recognition (ATR) method is developed based on the two-dimensional variational mode decomposition (2D-VMD). 2D-VMD decomposes original SAR images into multiscale components, which depict the time-frequency properties of the targets. The original image and its 2D-VMD components are highly correlated, so the multitask sparse representation is chosen to jointly represent them. According to the resulted reconstruction errors of different classes, the target label of test sample can be classified. The moving and stationary target acquisition and recognition (MSTAR) dataset is used to set up the standard operating condition (SOC) and several extended operating conditions (EOCs) including configuration variants, depression angle variances, noise corruption, and partial occlusion to test and validate the proposed method. The results confirm the effectiveness and robustness of the proposed method compared with several state-of-the-art SAR ATR references.

It is necessary to study the problem of vehicle routing in multidistribution centers to improve the speed, time, and cost thereof. It is preferable to use as few vehicles as possible to complete the delivery of goods and minimize the total mileage. With the development of artificial intelligence technology, machine learning is usually used to solve the problem of k shortest paths in multiple distribution centers. User needs are constantly changing; the iterative convergence speed of traditional machine learning methods is low and cannot meet the requirements of path planning in a big-data environment. Aiming at difficult problems in multipath planning, the parallel characteristics of traditional machine learning algorithms are fully exploited; k-means clustering and simulated annealing algorithms are improved through the distributed computing; and the multiple depot vehicle routing problem clustering analysis and path planning under the framework of Spark distributed computing are proposed. Through 30 simulation experiments on the TSPLIB dataset, the optimal solution is obtained with a 100% accuracy rate in problem solving. Experimental comparison and analysis show that the algorithm proposed in this article can solve the problem at least twice as fast as other parallel algorithms. This finding verifies that this method can effectively solve the multipath planning problem, thus greatly improving the quality and efficiency of path planning in large-scale logistics.

Building extraction from high-resolution aerial images is critical in geospatial applications such as telecommunications, dynamic urban monitoring, updating geographic databases, urban planning, disaster monitoring, and navigation. Automatic building extraction is a massive task because buildings in various places have varied spectral and geometric qualities. As a result, traditional image processing approaches are insufficient for autonomous building extraction from high-resolution aerial imaging applications. Automatic object extraction from high-resolution images has been achieved using semantic segmentation and deep learning models, which have become increasingly important in recent years. In this study, the U-Net model was used for building extraction, initially designed for biomedical image analysis. The encoder part of the U-Net model has been improved with ResNet50, VGG19, VGG16, DenseNet169, and Xception. However, three other models have been implemented to test the performance of the model studied: PSPNet, FPN, and LinkNet. The performance analysis through the intersection of union method has shown that U-Net with the VGG16 encoder presents the best results compared to the other models with a high IoU score of 83.06%. This research aims to examine the effectiveness of these four approaches for extracting buildings from high-resolution aerial data.