معالي هاشم حاكم العميدي

Abstract A deep learning powerful models of machine learning indicated better performance as precision and speed for images classification. The purpose of this paper is the detection of patients suspected of pneumonia and a novel coronavirus. Convolutional Neural Network (CNN) is utilized for features extract and it classifies, where CNN classify features into three classes are COVID-19, NORMAL, and PNEUMONIA. In CNN updating weights by CNN backpropagation and SGDM optimization algorithms in the training stage. The performance of CNN on the dataset is a combination between Chest X-Ray dataset (1583-NORMAL images and 4272-PNEUMONIA images) and COVID-19 dataset (126-images) for automatically anticipate whether a patient has COVID-19 or PNEUMONIA, where accuracy 94.31% and F1-Score 88.48% in case 60% training, 20% testing, and 20% validation. Keywords: Deep learning, convolutional neural network, and Coronavirus Disease (COVID-19). 2010 MSC: Primary 90C33; Secondary 26B25.

In late 2019, a virus appeared suddenly he claims Covid-19, which started in China and began to spread very widely around the world. And because of its effects, which are not limited to human life only, but rather in economic and social aspects, and because of the increase in daily injuries and significantly with the limited hospitals that cannot accommodate these large numbers, it is necessary to find an automatic and rapid detection method that limits the spread of the disease and its detection at an early stage in order to be treated more quickly. In this paper, deep learning was relied upon to create a CNN model to detect COVID-19 infected lungs using chest X-ray images. The base consists of a set of images taken of lungs infected with Covid-19 disease and normal lungs, as the CNN structure gave accuracy, Precision, Recall and F-Measure 100%

The problem is that e-commerce is not reliable and people are afraid to buy things and make transactions with commercial websites. Only some well-known companies have a known digital document from where it was issued. To solve this problem and to document any individual commercial website, We suggested to design a personal commercial website, it is a website to buy books and enable access to a domain by purchasing it from the digital ocean cloud, then a server is installed in this domain and then we connect it to the digital certification from the website (lets encrypt), where certain software are added to client (certbot, root) server assistant program, and we have linked the server and the certification together via Linux instructions and codes. The commercial website.(PHP3. PHP) has been tested in SSL laboratories and the results are the division of service sites into three categories: current, best current, and worst currently with details of categories, and what are the weaknesses.

ABSTRACT Classification is one of the applications of feed-forward Artificial Neural Network (ANN). Classification can map data to predefined classes or groups. It is referred to as a supervised learning, because before examining data the classes are always determined. Multi-Layer Perception, is a supervised neutral networks model that is use to train and test data to build a model. In this experiment. Multi-Layer Perception is used to train the Data set to produce a model to make prediction of classifying .After preparing the Mushrooms data for training, only 8124 of dataset instances used to be train. Software used to mining data in this project is Neural Connection Version 2.0. This report, generally explaining the Classification, Multi-Layer Preceptor, Back propagation, Mushrooms, and details on the mining activity done to the selected datasets, to determine whether Mushroom’s attribute is edible or Poison. KEYWORDS: Classification, Multi-Layer Preceptor, Back propagation, Mushrooms

Abstract: Clustering using Artificial Neural Network (ANN) is a technique used in data mining. In neural network models have two types: supervised neural networks and unsupervised neural networks. Kohonen Networks is an unsupervised neural networks that used to train and test historical data in order to build clustering a model. In this experiment, The main objective of the experiment is to cluster the Thyroid disease Data by using model of Kohonen Networks. In this experiment, neural connection software is used to run the clustering by Thyroid disease dataset supplied by Werner, which adopted from UCI Repository. Clustering techniques can be applied when we can't find class to predict but rather when we can divide the instances into natural groups.