Cloud providers frequently utilize two tightly coupled resource management strategies like task scheduling & data replication to boost the performance of the system generally, guaranteeing service level agreement (SLA) compliance, as well as protecting their own financial gain. An Improved Correlation strategy-based Task Scheduling and Data Replication in Cloud (ICTSDC) is what this work aims to give. The suggested system's primary phases are as follows: Management of replication and task scheduling. Initial job scheduling will be optimization-based and take into account goals such bottleneck value, migration cost, VM load, enhanced correlation, and replication, respectively. For this, a brand-new extended DMO algorithm called Self-adaptive Dwarf Mongoose Optimization (SADMO) is presented. In the replication management stage, the potential copies must first be identified based on the prior objective. The suggested SADMO model implements the optimization technique for replica placement throughout the replication management process. The outcomes of the ICTSDC technique are evaluated to other methods using a variety of metrics, like bottleneck value, migration cost, Virtual Machine (VM) load, improved correlation, as well as replication efficiency. A lower mean value of 0.324 is gained with the ICTSDC scheme for fitness.

ABSTRACT Applications like business basket analysis, digital service analytics, bio-informatics, and mobile commerce have greatly benefited from the information retrieval of significant features from massive databases for improved decision-making. Item set mining is used to find intriguing patterns in databases. Discovering item sets in an uncertain database is a tedious task. Only mathematical correlations between the elements in an item set are the exclusive subject of recurring item set mining research. The finding is direct to optimal. This article introduces an ant colony that maps the viable solution space to a directed graph with quadratic space complexity. The proposed model evaluates an uncertain transaction database's item set. Compared to the current methods, the findings demonstrate the importance of the proposed model.

Paddy disease recognition presents challenges in the agricultural industry, and existing algorithms struggle to accurately identify diseases in complex scenarios. In this paper, we propose a precise object detection framework to address the challenges of severe overlap, multi-disease detection, morphological irregularities, multi-scale object classification, and complex scenarios in real-world environments in paddy disease detection. The proposed model is based on an improved version of the DEtection TRansformer (Detr) algorithm. The enhanced network architecture fuses multi-scale features by adding a feature fusion module after the backbone network, which is able to retain more original information of the images and greatly improves the detection accuracy; the use of deformable attention module greatly reduces the computational complexity of the model. To evaluate the PDN, a dedicated paddy disease detection dataset with 1200 images is created. Experimental results demonstrate that the proposed model obtained a precision value of 100%, a recall value of 89.3%, F1-score of 94.3%, and a mean average precision (mAP) value of 60.2%. The model outperforms the existing state-of-the-art detection models in detection accuracy.

ABSTRACT This research presents an advanced approach to enhance disease diagnosis using imbalanced medical datasets. Feature selection techniques, LASSO and Relief, are applied to identify relevant features from the UCI dataset and missing values are handled appropriately. To address the class imbalance, SMOTEENN is used, creating a new combined dataset with selected features. Three deep learning models, FNNs, LSTMs, and GBMs, are employed and trained on the combined dataset, achieving remarkable accuracy (1.0). Evaluating the models on LASSO and Relief datasets independently, FNN/MLP obtains perfect accuracy, GBM performs well (0.9888 on LASSO and 1.0 on Relief), and LSTM shows good results (0.9663 on LASSO and 1.0 on Relief). This study demonstrates the effectiveness of combining LASSO and Relief for feature selection and highlights the impact of SMOTEENN on model performance. The achieved accuracy with all models on the combined dataset showcases deep learning's potential for accurate disease diagnosis even with imbalanced data, offering promising insights for robust medical diagnosis systems.

A model with high accuracy of machine failure prediction is important for any machine life cycle. In this paper, a prediction model based on machine learning methods is proposed. The used method is a combination of machine learning algorithms and techniques. The machine learning algorithm is a data mining technique that has been widely used as a prediction model for classifying problems. Five algorithms have been tested including JRIP, logistic, KStar, Bayes network and decision table machine learning. The evaluation process is done by applying the algorithms on a predictive dataset using different performance measures. In the proposed model, the feature selection and voting techniques are used and applied in the classification process for each classifier. From the comparison of the result, the feature selection shows the best performance result. Paired t-test evaluation measures were considered to confirm our conclusion. The best accuracy result among the five classifiers shows that joint reserve intelligence classifier can be used to predict the failure with an accuracy high as 0.983. Applying classifier subset evaluation using the JRIP classifier can enhance the accuracy result to be 0.985. The finding shows that the proposed model improves the results of the classifiers.

In this study, the cloud computing platform is equipped with a hybrid multi-objective meta-heuristic optimization-based load balancing model. Physical Machine (PM) allocates a specific virtual machine (VM) depending on multiple criteria, such as the amount of memory used, migration expenses, power usage, and the load balancing settings, upon receiving a request to handle a cloud user's duties (‘Response time, Turnaround time, and Server load’). Additionally, the optimal virtual machine (VM) is chosen for efficient load balancing by utilizing the recently proposed hybrid optimization approach. The Cat and Mouse-Based Optimizer (CMBO) and Standard Dingo Optimizer (DXO) are conceptually blended together to get the proposed hybridization method known as Dingo Customized Cat mouse Optimization (DCCO). The developed method achieves the lowest server load in cloud environment 1 is 33.3%, 40%, 42.3%, 40.2%, 36.8%, 42.5%, 50%, 40.2%, 39.2% improved over MOA, ABC, CSO, SSO, SSA, ACSO, SMO, CMBO, BOA, DOX, and FF-PSO, respectively. Finally, the projected DCCO model has been evaluated in terms of makespan, memory usage, migration cost, response time, usage of power server load, turnaround time, throughput, and convergence. ABBREVIATION: CDC, cloud data center; CMODLB, Clustering-based Multiple Objective Dynamic Load Balancing As A Load Balancing; CSP, Cloud service providers; CSSA, Chaotic Squirrel Search Algorithm; DA, Dragonfly Algorithm; ED, Euclidean Distance; EDA-GA, Estimation Of Distribution Algorithm And GA; FF, FireFly algorithm; GA, Genetic Algorithm; HHO, Harris Hawk Optimization; IaaS, Infrastructure-as-a-Service; MGWO, Modified Mean Grey Wolf Optimization Algorithm; MMHHO, Mantaray modified multi-objective Harris Hawk optimization; MRFO, Manta Ray Forging Optimization; PaaS, Platform-as-a-Service; PM, Physical Machine; PSO, Particle Swarm Optimization; SaaS, Software-as-a-Service; SAW, Sample additive weighting; SLA-LB, Service Level Agreement-Based Load Balancing; TBTS, Threshold-Based Task Scheduling Algorithm; TS, Task Scheduling

Block encryption algorithms are among the most preferred applications in cryptographic systems. Block ciphers should have accomplished some requirements for a secure communication system. They should be evaluated in terms of cryptanalysis methods for widespread usage. The aim of this paper is to introduce a new secure and fast block encryption algorithm for images. For this purpose, a new block cipher, which offers an innovative encryption structure for key generation systems and can use S-boxes with different methods, is proposed. A Dynamic S-Box is used in the algorithm for both substitution and key generation purposes. Linear and differential cryptanalysis methods were performed successfully. UACI and NPCR tests show that the proposed symmetric block cipher algorithm is compatible with image encryption systems. The 512-bit key length provides the highest security for block encryption. Additionally, information entropy test, correlation coefficients, mean-squared error, and peak signal-to-noise ratio analyses were concluded successfully. The novelty of the paper is building a cryptanalysis attack-resistant block cipher algorithm that presents a lightweight cryptographic solution for image encryption systems.

Because of the rapid advancement of technology over the last several years, the number of internet users is growing at an exponential rate, and as a result, email communication has become popular as a means of exchanging information over the internet. Sending data and communicating with peers via email is the most cost-effective method. These email services also cause problems for users by sending electronic junk mail, often known as spam mail. Spam email is a privacy concern that is linked to a slew of commercial and dangerous websites, causing phishing, virus distribution, and a slew of other problems. This study examines several aspects that have been used for email spam classification, as well as offering an overview of a handful of classifiers or algorithms that have been successfully evaluated, as well as exploratory data analysis. The proposed email spam classifier uses three parallel layers of machine learning and deep learning techniques, followed by a decision function to determine whether or not the emails are spam. During testing, it was found that the proposed classifier beats similar systems on the standard dataset with an accuracy of 98.4%.

Background initialization is used in video processing applications to extract a scene without the foreground scene. In recent times, the issue of background initialization has gained researchers’ attention in different fields such as video surveillance, video segmentation, computational photography, and so on. The initialization of the background is affected due to different complex dissimilarities such as shadow, intermittent movement, illumination, camera jitter, and clutter. To overcome the aforementioned issues, this paper proposes a decomposition using the combination of the Singular Value Decomposition (SVD) and Robust Principal Component Analysis (RPCA) for Singular Spectrum Analysis (SSA) to perform an effective background initialization. The incorporation of RPCA in SVD is used to overcome the issues related to non-Gaussian noise and it also uses an effective structural knowledge of the video input i.e. sparse and low rank matrix which improves the Peak-Signal-to-Noise-Ratio (PSNR) of the background image. The SBI dataset was used to analyze the performances of SSA-SVDRPCA. The SSA-SVDRPCA is analyzed using MultiScale Structural Similarity Index (MSSSIM), Average gray-level error (AGE), Percentage of clustered error pixels (pCEPS), Percentage of error pixels (pEPs), and PSNR. The existing approaches such as Background Initialization Singular Spectrum Analysis (BISSA) and Quaternion-based Dynamic Mode Decomposition (Q-DMD) are used to compare with the SSA-SVDRPCA method. The PSNR of the SSA-SVDRPCA for Board class is 30.39 dB which is higher than the BISSA and Q-DMD.

Technological advancements have resulted in an exponential increase in the use of online social networks (OSNs) worldwide. While online social networks provide a great communication medium, they also increase the user's exposure to life-threatening situations such as suicide, eating disorder, cybercrime, compulsive behavior, anxiety, and depression. To tackle the issue of cyberbullying, most existing literature focuses on developing approaches to identifying factors and understanding the textual factors associated with cyberbullying. While most of these approaches have brought great success in cyberbullying research, data availability needed to develop model detection remains a challenge in the research space. This paper conducts a comprehensive literature review to provide an understanding of cyberbullying detection.