Portal of UHD Academic Journals
Not a member yet
496 research outputs found
Sort by
Kurdish Speech to Text Recognition System Based on Deep Convolutional-recurrent Neural Networks
Get PDFIn recent years, deep learning has had enormous success in speech recognition and natural language processing. In other languages, recent progress in speech recognition has been quite promising, but the Kurdish language has not seen comparable development. There are extremely few research papers on Kurdish speech recognition. In this paper, investigated Gated Recurrent Units (GRUs) which is one of the popular RNN models to recognize individual Kurdish words, and propose a very simplified deep-learning architecture to get more efficient and high accuracy model. The proposed model consists of a combination of CNN and GRU layers. The Kurdish Sorani Speech KSS dataset was created for the speech recognition system, as its 18799 sound files for 500 formal Kurdish words. Finally, the model proposed was trained with collected data and yielded over %96 accuracy. The combination of CNN an RNN (GURs) for speech recognition achieved superior performance compared to the other feed-forward deep neural network models and other statistical methods
COVID-19 Disease Detection Based on Machine Learning and Chest X-Ray Images
Get PDFDue to increasing population, automated illness identification has become a critical problem in medical research. An automated illness detection framework aids physicians in disease diagnosis by providing precise, consistent, and quick findings, as well as lowering the mortality rate. Coronavirus (COVID-19) has expanded worldwide and is now one of the most severe and acute disorders. To avoid COVID-19 from spreading, making an automatic detection system based on X-ray chest pictures ought to be the quickest diagnostic alternative. The goal of this research is to come up with the best model for detecting COVID-19 diagnosis with the greatest accuracy. Therefore, four models, Convolutional Neural Networks, Residual Network 50, Visual Geometry Group 16 (VGG16), and VGG19, have been evaluated using the same images preprocessing method. In this study, performance metrics include accuracy, precision, recall, and F1 scores are used for evaluating proposed method. According to our findings, the VGG16 model is a viable candidate for detecting COVID-19 instances, because it has highest accuracy; in result overall accuracy of 98.44% in training phase, 98.05% invalidation phase and 96.05% in testing phase is obtained. The results of other performance measurements are shown in the result section, demonstrating that the majority of the approaches are more than 90% accurate. Based on these results, radiologists may find the proposed VGG16 model to be an intriguing and a helpful tool for detecting and diagnosing COVID-19 patients quickly
Newly Simple Quantitative Determination of Montelukast Sodium by Ultraviolet-Spectrophotometry
Get PDFMontelukast sodium is well known pharmaceutically for its action as leukotriene antagonist and reliving symptoms associated with asthma is available in the market as tablet, chewable tablet, and powder. The aim of this study was to develop newly simple selective ultraviolet spectrophotometry (UV) method for daily routine analysis of quality control department. The UV method was developed with wavelength at 287.0 nm. This newly developed method was effectively applied to tablet dosage form of the motelukast sodium follow the Beer’s Lamberts at range 2.5–50 μg/mL. The validated parameters were carryout such as linearity, accuracy, precision, and specificity. The result of validation statistically studied and found to be satisfactory
Malicious URL Detection Using Decision Tree-based Lexical Features Selection and Multilayer Perceptron Model
Get PDFNetwork information security risks multiply and become more dangerous. Hackers today generally target end-to-end technology and take advantage of human weaknesses. Furthermore, hackers take advantage of technology weaknesses by applying various methods to attack. Nowadays, one of the greatest dangers to the modern digital world is malicious URLs, and stopping them is one of the biggest challenges in the field of cyber security. Detecting harmful URLs using machine learning and deep learning algorithms have been the subject of various academic papers. However, time and accuracy are the two biggest challenges of these tools. This paper proposes a multilayer perceptron (MLP) model that utilizes two significant aspects to make it more practical, lightweight, and fast: Using only lexical features and a decision tree (DT) algorithm to select the best relevant subset of features. The effectiveness of the experimental outcomes is evaluated in terms of time, accuracy, and error reduction. The results show that a MLP model using 35 features could achieve an accuracy of 94.51% utilizing only URL lexical features. Furthermore, the model is improved in time after applying the DT as feature selection with a slight improvement in accuracy and loss
Classification of Acute Lymphoblastic Leukemia through the Fusion of Local Descriptors
Get PDFLeukemia is characterized by an abnormal proliferation of leukocytes in the bone marrow and blood, which is usually detected by pathologists using a microscope to examine a blood smear. Leukemia identification and diagnosis in advance are a trending topic in medical applications for decreasing the death toll of individuals with Acute Lymphoblastic Leukemia (ALL). It is critical to analyze the white blood cells for the identification of ALL for which the blood smear images are utilized. This paper discusses and presents a micro-pattern descriptor, called Local Directional Number Pattern along with Multi-scale Weber Local Descriptor for feature extraction task to determine cancerous and noncancerous blood cells. A balanced dataset with 260 blood smear images from the ALL-IDB2 dataset was used as training data. Consequently, a proposed model was constructed by applying different individual and combined feature extraction methods, and fed into the machine learning classifiers (Decision Tree, Ensemble, K-Nearest Neighbors, Naïve Bayes, and Random Forest) to determine cancerous and noncancerous blood cells. Experimental results indicate that the developed feature fusion technique assured a reasonable performance compared to other existing works with a testing average accuracy of 97.69 ± 1.83% using Ensemble classifier
Log File Analysis Based on Machine Learning: A Survey: Survey
No full textIn the past few years, software monitoring and log analysis become very interesting topics because it supports developers during software developing, identify problems with software systems and solving some of security issues. A log file is a computer-generated data file which provides information on use patterns, activities, and processes occurring within an operating system, application, server, or other devices. The traditional manual log inspection and analysis became impractical and almost impossible due logs’ nature as unstructured, to address this challenge, Machine Learning (ML) is regarded as a reliable solution to analyze log files automatically. This survey tries to explore the existing ML approaches and techniques which are utilized in analyzing log file types. It retrieves and presents the existing relevant studies from different scholar databases, then delivers a detailed comparison among them. It also thoroughly reviews utilized ML techniques in inspecting log files and defines the existing challenges and obstacles for this domain that requires further improvements
Photosynthetic Pigments and Stomata Characteristics of Cowpea (Vigna sinensis savi) under the Effect of X-Ray Radiation
Get PDFThis study was conducted in the field and laboratories of the faculty of science and health-Koya university by exposing the seeds of cowpea plant (Vigna sinensis Savi) var. California black-eye to X-ray radiation in two different locations (In target or 30 cm out of target) inside the radiation chamber, for four different exposure times (0, 5, 10, or 20 min), to study the effect on some characteristics of seedling components. Results show that the exposure location to X-ray had non-significant effects on cowpea leaves content of photosynthetic pigments, whereas each of time of exposure with interaction between location and time of exposure had significant effects on chlorophyll a, total chlorophylls, and total carotenoids pigments. Regarding the X-ray effects on stomata characteristics, the results detect that there were non-significant differences between the location of exposure on stomata number on abaxial leaves surfaces and stomata length on adaxial leaves surfaces, whereas a significant effects on number of stomata on the adaxial leaves surfaces, abaxial stomata length, abaxial, and adaxial stomata width were detect. Exposing cowpea seeds to X-ray radiation in the target of the radiation source increased significantly stem and leave dry matter percent compared with the one out of the target location, whereas increasing the time of exposure decreased the percent of dry matter of stem and leaves. It is concluded that exposing cowpea seeds to X-ray leads to changes in photosynthetic pigments, stomata characteristics, and plant dry matter content
Adaptive Filter based on Absolute Average Error Adaptive Algorithm for Modeling System
Get PDFAdaptive identification of the bandpass finite impulse response (FIR) filtering system is proposed through this paper using variable step-size least mean square (VSS-LMS) algorithm called absolute average error-based adjusted step-size LMS as an adapted algorithm. This algorithm used to design an adaptive FIR filter by calculating the absolute averaged value for the recently assessed error with the previous one. Then, the step size has been attuned accordingly with consideration of the slick transition of the step size from bigger to smaller to score an achievement through high convergence rate and low steady-state misadjustment over the other algorithms used for the same purpose. The simulation results through the computer demonstrate remarkable performance compared to the traditional algorithm of LMS and another VSS-LMS algorithm (normalized LMS) which used in this paper for the designed filter. The powerful of the filter has been served in the identification system, bandpass filter has been chosen to be identified in the proposed adaptive system identification. It reports conceivable enhancements in the modeling system concerning the time of convergence, which is well-defined as a fast and steady-state adjustment defined with a low level. The designed filter identified the indefinite system with less than 10 samples; meanwhile, other algorithms were taking more than 20 samples for identification. Alongside the fine behavior of preserving the tradeoff between miss adjustment and the capability of tracking, the fewer calculations and computations regarding the algorithm requirement make the applied real-time striking
Semantic Web Recommender System over Different Operating Platforms
Get PDFSemantic-Web Recommender System (SWRS) evaluation over different operating systems (OSs) used to facilitate and improve human electronic recommendation management (HERM). The HERM is address the needs of user and dataset of movie in our proposed system through internetworking means which increase the speed of automated recommendation and enhance the goodness of SWRS and services also electronically to select right movies-title to user demand. Furthermore, it will be a benefit for selection a right favor by user for right selection from (i.e., 3000 records in dataset of movie-Lens) in the backend. There are a direct relation between time-consume of selection movie-title, also the time-consume, and accuracy. The two-mentioned parameters, namely, time-consume and accuracy over two different operation system (OSs) which designed by web technology Python. In our research, SWR system is proposed; it is provide with some recommendation methods. The system designed and improved using content-based algorithm (CBA). Investigational results indicate that the developed algorithm technique confident a reasonable performance such as accuracy and time consuming compared to other existing works with a testing average accuracy of 85.63 for windows and 88.35 for Linux operating system. In conclusion, SWRS investigated on two different operating platforms and could be seen that the Linux is faster than windows in accuracy and time consuming
A Secure Medical Image Transmission System Based On 2D Logistic Map and Diffie-Hellman Key Exchange Mechanisms
Get PDFWith the tremendous growth of searchable visual media, the content-based medical image retrieval and computer-aided diagnosis systems have become popular in recent years to improve knowledge and provide facilities for radiologists. Medical images transferred throughout public networks demand a mechanism that guarantees image privacy, ownership and source of origin reliability, and image integrity verification. For this reason, secure image retrieval and diagnosis scheme have been given considerable interest due to users’ security concerns. This work proposed a secure framework based on a two-dimensional (2D) chaotic map with Diffie–Hellman key exchange protocols to ensure patient information privacy and security. Consequently, from a security and protection perspective, the objective is to provide a privacy procedure for medical image retrieval systems through image encryption technique combined with a secure key exchange procedure to minimize the possibility of secret key interception by an unauthorized person. Simulation results and security analysis show that the suggested technique could protect images with minimal time complexity and a high level of security while also resisting numerous attacks