Iraqi Journal for Computer Science and Mathematics

To solve the problem of large recognition errors in traditional attack information identification methods, we propose a machine learning (ML)-based identification method for electric power Internet attack information. Based on the Internet attack information, an Internet attack information model is constructed, the identification principle of the power Internet attack information is analysed based on ML, hash fixing is conducted to ensure that the same attack information will be assigned to the same thread and that the deviation generated by noise
can be avoided so that the real-time lossless processing of the power Internet attack information can be ensured. The vulnerability adjacency matrix is constructed, and the vulnerability is quantitatively evaluated to complete the design of the optimal identification scheme for power Internet attack information. The experimental results show that the identification accuracy of the method can reach 98%, which can effectively reduce the risk of power Internet network attacks and ensure the safe and stable operation of the network.

Over the last decade, the chaotic behaviors of dynamical systems have been extensively explored. Recently, discovering or developing a 2D system of ordinary differential equations (ODEs) capable of exhibiting
chaotic dynamical behaviors is an attractive research topic. In this study, a chaotic system with a 2D system of
nonsmooth ODEs has been developed. This system is can exhibit chaotic dynamical behaviors. Its main dynamical behaviors, including time-series trajectories, phase portraits of attractors, and equilibria and their stability, have been investigated. The developed system has been verified by an excessive variety of fascinating chaotic behaviors, such as chaotic attractor, symmetry, sensitivity to initial conditions (ICs), fractal dimension, autocorrelation, power spectrum, Lyapunov exponent, and bifurcation diagram. Analytical and numerical simulations are used to study the dynamical behaviors of such a system. The developed system has extreme sensitivity to ICs, a fractal dimension of more than 1.8 and less than 2.05, an autocorrelation fluctuating randomly about an average of zero, a broadband power spectrum, and one positive Lyapunov exponent. The obtained numerical simulation results have proven the capability of the developed 2D system for exciting chaotic dynamical behaviors.

The main aim of this study is to create a new type of topology called “ -simple extension” and investigate its properties. We introduce a new definition for -open sets and consider this aspect as the basis of our main definition. Furthermore, we investigate the properties of the proposed concept to allow us to provide new examples of explicit descriptions of topological spaces and certain types of -covering for topological spaces, such as ( -Lindelof and -paracompact spaces). The use of the tool offers important results for topological spaces. Other findings related to the proposed approach have also been identified.

The influx of data in bioinformatics is primarily in the form of DNA, RNA, and protein sequences. This condition places a significant burden on scientists and computers. Some genomics studies depend on clustering techniques to group similarly expressed genes into one cluster. Clustering is a type of unsupervised learning that can be used to divide unknown cluster data into clusters. The k-means and fuzzy c-means (FCM) algorithms are examples of algorithms that can be used for clustering. Consequently, clustering is a common approach that divides an input space into several homogeneous zones; it can be achieved using a variety of algorithms. This study used three models to cluster a brain tumor dataset. The first model uses FCM, which is used to cluster genes. FCM allows an object to belong to two or more clusters with a membership grade between zero and one and the sum of belonging to all clusters of each gene is equal to one. This paradigm is useful when dealing with microarray data. The total time required to implement the first model is 22.2589 s. The second model combines FCM and particle swarm optimization (PSO) to obtain better results. The hybrid algorithm, i.e., FCM–PSO, uses the DB index as objective function. The experimental results show that the proposed hybrid FCM–PSO method is effective. The total time of implementation of this model is 89.6087 s. The third model combines FCM with a genetic algorithm (GA) to obtain better results. This hybrid algorithm also uses the DB index as objective function. The experimental results show that the proposed hybrid FCM–GA method is effective. Its total time of implementation is 50.8021 s. In addition, this study uses cluster validity indexes to determine the best partitioning for the underlying data. Internal validity indexes include the Jaccard, Davies Bouldin, Dunn, Xie–Beni, and silhouette. Meanwhile, external validity indexes include Minkowski, adjusted Rand, and percentage of correctly categorized pairings. Experiments conducted on brain tumor gene expression data demonstrate that the techniques used in this study outperform traditional models in terms of stability and biological significance.

This study proposes a two-stage shrinkage Bayesian estimation of the shape parameter of Pareto
distribution. Additional information from the past and considered presently in new estimation processes has been receiving considerable attention in the last few decades, especially when a sample unit is costly or difficult to obtain. The proposed two-stage pooling estimation procedure assumes that the prior knowledge of ? can take the form of an initial estimate ?0 of ?. The expressions for bias, bias ratio, mean square error, expected sample size, and relative efficiency are derived based on the two regions of R1 and R2. Certain values of the constants are considered, and the R language is used for statistical programming. The numerical results and conclusion suggest that the proposed estimators have higher relative efficiency compared with the classical Bayesian estimator with respect to a guess value. The effective region of the estimator dependent on R2 is better than that of the estimator dependent dependent on R1.

A distributed denial-of-service (DDoS) attack attempts to prevent people from accessing a server. A
website may become inaccessible due to a DDoS attack because the server is inundated with fake requests and cannot handle real ones. A DDoS attack affects a large number of computers. Attackers employ a zombie network, which is a collection of infected machines on which the attacker has hidden the denial-of-service attacking application to carry out a DDoS attack. The MATLAB 2018a simulator was used in this study for training. Additionally, during design, the knowledge discovery dataset (KDD) was cleaned and the values of attacks were incorporated. A neural network model was subsequently developed, and the KDD was trained using a recursive artificial neural network. This network was developed using five distinct training algorithms: 1) Fletcher–Powell conjugate gradient, 2) Polak–Ribiére conjugate gradient of, 3) resilient backpropagation, 4) gradient conjugation with Powell/Beale restarts, and 5) gradient descent algorithm with variable learning rate. The artificial neural network toolset in MATLAB was used to investigate the detection of DDoS attacks. The conjugate gradient with Powell/Beale restart algorithm had a success rate of 99.9% and a training time of 00:53. This inquiry uses the KDD-CUP99 dataset. Has a better level of accuracy, according to the results.

Users of social media can consume a wide range of subjects and information, including pornography. Pornography usage as well as the difficulties linked with this sort of content have increased over time, particularly among teens. Now, another sort of pornographic content is popular: child pornography (CP). Controlling online CP has always been a difficult task for the international community. The introduction, growth, and use of information and communication technologies coincide with an increase in illegal activities. In terms of cyberspace, children’s huge online presence as well as the emergence of CP as a business compel all governments to enact tough legislation and unite worldwide to combat this issue. Is social media assisting in the dissemination of this content? To understand this issue further, a study conducted in 2021 using Telegram data on the consumption of CP followed by a series of analyses show Telegram’s potential effect on the spread and consumption of this type of content.

The need for network intrusion detection systems (NIDS) to protect against different attacks grows as the scale of cyber attacks increases. The main areas of cyber attack research are its detection and prevention. Traditional machine learning (ML) algorithms with low accuracy are used by the current NIDS, but it is not suitable for newer anonymous cyber attacks. In this paper, an NIDS model with ensemble ML methods, which can detect and prevent different types of attacks compared with traditional ML methods, is proposed. Our specific system detects known attacks and blocks unknown attacks. The selected system uses four different machine learning methods, including data processing techniques for data preprocessing and data labeling. The entire NSL-KDD database is used to evaluate the performance of various ML classifiers based on different parameters. The simulation analysis shows that the developed NIDS system is better than the existing single ML methods. The detection accuracy rate of intrusion detection system (IDS) is increased by the model, which is essential for NIDS.

GANs (generative opposing networks) are a technique for learning deep representations in the absence of a large amount of annotated training data. This is accomplished through the use of a competitive technique that employs two networks to generate background signals. GANs can use learned representations for a variety of applications, including image synthesis, semantic imaging, style transfer, super magnification, and segmentation. Images can be utilized in a variety of ways. Generative Adversarial Networks (GANs) are a unique class that has recently received a lot of interest due to the popularity of deep generative models. GANs implicitly distribute complex and high-resolution images, sounds, and data. However, due to inadvertently built network architecture, objective function usage, and optimization algorithm selection, significant difficulties such as mode collapse, inconsistencies, and instability develop while training GANs. In this paper, we conduct a thorough examination of the developments in GANs design and optimization strategies presented to address GANs difficulties. We provide intriguing study possibilities in this rapidly evolving area. GANs are a popular study topic due to their ability to generate synthetic data and the benefits of representations that can be understood regardless of the application. While various reviews for GANs in the image processing arena have been undertaken to date, none have focused on the review of GANs in multi-disciplinary domains. As a result, the utilization of GAN in interdisciplinary applications fields and its implementation issues were investigated in this survey by doing a thorough search for journal/research article connected to GAN.

Yesterday’s “Big Data” is today’s “data.” As technology advances, new difficulties and new solutions emerge. In recent years, as a result of the development of Internet of Things (IoT) applications, the area of Data Mining has been confronted with the difficulty of analyzing and interpreting data streams in real time and at a high data throughput. This situation is known as the velocity element of big data. The rapid advancement of technology has come with an increased use of social media, computer networks, cloud computing, and the IoT. Experiments in the laboratory also generate a large quantity of data, which must be gathered, handled, and evaluated. This massive amount of data is referred to as “Big Data.” Analysts have seen an upsurge in data including valuable and worthless elements. In extracting usable information, data warehouses struggle to keep up with the rising volume of data collected. This article provides an overview of big data architecture and platforms, tools for data stream processing, and examples of implementations. Streaming computing is the focus of our project, which is building a data stream management system to deliver large-scale, cost-effective big data services. Owing to this study, the feasibility of large-scale data processing for distributed, real-time computing is improved even when the systems are overwhelmed.

Due to the increasing competitiveness in telecom’s market, it has now become more necessary for operators to start building personal relationship with customers for targeted retention strategies. Achieving this goal requires the development of an effective churn prediction model that will solve the problem of churn misclassification, which is persistent in current churn prediction models. With several existing segment-oriented churn prediction models failing to harness the power of associative networking provided by telecoms users, churn prediction accuracy remains unguaranteed while targeted decision support is not enhanced. Here, the research introduced the Customer’s Influence Degree (I) to the existing Recency, Frequency, and Monetary (RFM) values as an additional predictive factor, towards determining the churn class of a customer. The essence is to utilise the socio-transactional affinities of customers’ direct dependent to targeted communication nodes through customers RFM analysis to determine the dominance of a customer in the community. The newly introduced predictive factor helped to minimise churn misclassification rate through appropriate reclassification of customers who were wrongly classified as churner or non-churner when using the existing RFM churn scores only.

Python is an amazing language for large and complicated programming ventures. This language uses
codes that are simple to use or modify for any software programmer. In this paper, one annoying issue during the
project expansion was described. Once a package is updated to the new version, this package is no longer compatible
with previous projects, causing trouble when working on the program. To solve this problem, Python supports the
programmer with a function called virtual environment. This function can construct an independent environment for
any venture to avoid conflict. Various steps have been explained to create a virtual environment in Python. Different
examples have then shown the benefits of Python language to building shapes, animations, and interactions.

In this study, various organizations that have participated in several road path-detecting experiments are analyzed. However, the majority of techniques rely on attributes or form models built by humans to identify sections of the path. In this paper, a suggestion was made regarding a road path recognition structure that is dependent on a deep convolutional neural network. A tiny neural network has been developed to perform feature extraction to a massive collection of photographs to extract the suitable path feature. The parameters obtained from the model of the route classification network are utilized in the process of establishing the parameters of the layers that constitute the path detection network. The deep convolutional path discovery network’s production is pixel-based and focuses on the identification of path types and positions. To train it, a detection failure job is provided. Failure in path classification and regression are the two components that make up a planned detection failure function. Instead of laborious postprocessing, a straightforward solution to the problem of route marking can be found using observed path pixels in conjunction with a consensus of random examples. According to the findings of the experiments, the classification precision of the network for classifying every kind is higher than 98.3%. The simulation that was trained using the suggested detection failure function is capable of achieving an accuracy of detection that is 85.5% over a total of 30 distinct scenarios on the road.

Most of the world languages have made strides in analyzing and classifying texts electronically; hence, the use of electronic text has become a great alternative to manual classification as it reduces time, cost, and difficulty. However, in the Arabic language, electronic analysis has not progressed due to several limitations faced by researchers in this field, such as the complexity of the Arabic language, the lack of related research, as well as the use of the classical Arabic language. In addition, Arabic poetry has other limitations, such as the use of a system that uses a single activation function. In this research, a new method was developed for the classification of the classical Arabic poetry and Al-hur poetry. This new approach is based on features that indicate the type of poetry. Pre-processing of
some data is important in this new approach as it helps increase the accuracy of classification