One of the most promising techniques used in various sciences is deep neural networks (DNNs). A special type DNN called a convolutional network (CNN) consists several layers, each preceded by an activation function and pooling layer. The feature map previous layer sampled (that seems to be important layer) create new with condensed resolution. This significantly reduces spatial dimension input. It always accomplished two main goals. As first step, it number parameters or weights minimize...

Cloud computing is considered to be the best technique for storing data online instead of using a hard drive. It includes three different types services that are provided remote users via Internet. offers its end variety options, such as cost savings, access resources and performance, but number in cloud grows, so does likelihood an attack. Various researchers have researched many solutions prevent these attacks. One ways detect attack through Intrusion Detection System. This article will...

Semi-supervised learning has been successfully connected in the research fields of machine such as data mining and dynamic analysis. Imbalance class is one most challenging issues for classification. In recent years, core focal point numerous researchers on classification multi-class imbalanced datasets. this paper, we proposed semi-supervised deep Fuzzy C-mean clustering (DFCM-MC). our word "Deep" used to show how decomposition strategy applied deeply, first, decomposes original into...

The amount of mobile applications is increasing rapidly, and it difficult for software developers to identify the numerous key factors that affect their rating performance. This study presents a machine-learning framework improve decisions in adding new features enhancing overall A dataset app attributes from Apple AppStore was used, exploiting NLP techniques preprocess textual information develop an Enhanced Random Forest (ERF) assess forecast ratings multifunctional apps investigate...

To determine the frequency and types of dermatological conditions including sexually-transmitted infections (STIs) within lesbian, gay, bisexual, transgender, queer (LGBTQ) communities. A cross-sectional survey. Place Duration Study: Department Dermatology, Services Hospital, Lahore, Pakistan, from 19 April to 20 May 2023. The survey was conducted on members LGBTQ community registered with Fountain House, aged 18 years above. Data demographics, sexual orientation, hormone use,...

Software fault prediction is a consequential research area in software quality promise. In this paper, we propose semi-supervised deep fuzzy C-mean (DFCM) clustering for prediction, which the cumulation of DFCM and feature compression techniques. Deep utilized feature-based multi clusters unlabeled labeled data sets along with their classes. our approach, training model, simultaneously deal unsupervised supervised to exploit obnubilated information from support construction precise model. We...

Defect detection is very important for guaranteeing the quality and pricing of fabric. A considerable amount fabric discarded as waste because defects, leading to substantial annual losses. While manual inspection has traditionally been norm detection, adopting an automatic defect scheme based on a deep learning model offers timely efficient solution assessing quality. In real-time manufacturing scenarios, datasets lack high-quality, precisely positioned images. Moreover, both plain printed...

Software fault prediction is the very important research topic for software quality assurance. Data-driven approaches provide robust mechanisms to deal with prediction. However, performance of model highly depends on data set. Many sets suffer from problem class imbalance. In this regard, undersampling a popular pre-processing method in dealing imbalance problem; easy ensemble presents approach achieve high classification rate and address biases toward majority samples. not only issue that...

The multi-objective grasshopper optimization algorithm (MOGOA) is a relatively new inspired by the collective behavior of grasshoppers, which aims to solve problems in IoT applications. In order enhance its performance and improve global convergence speed, integrates simulated annealing (SA). Simulated metaheuristic that commonly used search capability algorithms. case MOGOA, integrated employing symmetric perturbation control movement grasshoppers. This helps effectively balancing...

Data security assurance is crucial due to the increasing prevalence of cloud computing and its widespread use across different industries, especially in light growing number cybersecurity threats.A major everpresent threat Ransomware-as-a-Service (RaaS) assaults, which enable even individuals with minimal technical knowledge conduct ransomware operations.This study provides a new approach for RaaS attack detection uses an ensemble deep learning models.For this purpose, network intrusion...

Owing to their increasing rate of discovery as incidental findings, the characterization adrenal masses is an important diagnostic problem which frequently challenges clinician's skill.The results dehydroepiandrosterone sulfate (DHEAS) measurement were evaluated in a consecutive series 107 patients with mass (39 men, 68 women aged 15-81 years, median 56 years). DHEAS levels observed categorized reduced, normal or elevated according sex- and age-adjusted reference ranges obtained by measuring...

Software fault prediction is a very consequent research topic for software quality assurance. The performance of model depends on the features that are used to train it. Redundant and irrelevant can hinder classification model. In this paper, we propose an empirical study two-stage data pre-processing technique models. first stage, novel semi-supervised deep Fuzzy C-Mean (DFCM) clustering-based feature extraction proposed create new by utilizing multi-clusters unlabeled labeled sets tends...

SUMMARY Objective In this study, we report acute blood transfusion reactions at our hospital, compare analysis with the reported data and identify areas for improvement. Background Haemovigilance programmes have been implemented in many countries, adverse events associated are published their annual reports. Pakistan has no current established programme. Material Methods A cross‐sectional study was conducted, all to bank from January 2014 March 2016 were included. An response patient,...

While deep learning (DL) has proven to be a powerful paradigm for the classification of large-scale big image data set. Deep network such as CNN requires large number labeled samples training network. However, often times are difficult, expensive, and time-consuming. In this paper, we propose semi-supervised approach fused fuzzy-rough C-mean clustering with convolutional neural networks (CNNs) knowledge from simultaneously intra-model inter-model relationships forming final representation...

Researchers have created cryptography algorithms that encrypt data using a public or private key to secure it from intruders. It is insufficient protect the by such key. No research article has identified an algorithm capable of protecting both and associated key, nor any mechanism been developed determine whether access permissible impermissible based on authentication This paper presents WEDEx-Kerberotic Framework for protection, in which user-defined firstly converted cipher “Secure Words...

In today's world, businesses and individuals alike rely on cloud computing for storing accessing data. However, with the growing number of cyber threats, it is crucial to prioritize security. Despite various algorithms designed prevent cyber-attacks, attackers have developed advanced tactics bypass security measures. That's why essential understand three primary platforms how they work together create a seamless environment. Cloud cryptography highly effective method encrypting data enabling...

Deep learning has been well-known for a couple of years, and it indicates incredible possibilities unsupervised representations with the clustering algorithm. The forms Convolution Neural Networks (CNN) are now state-of-the-art many recognition tasks. However, perpetual incrementation digital images, there exist more redundant, irrelevant, noisy samples which cause CNN running to gradually decrease, its accuracy decreases concurrently. To conquer these issues, we proposed an effective method...