total accuracy
Recently Published Documents


TOTAL DOCUMENTS

74
(FIVE YEARS 45)

H-INDEX

9
(FIVE YEARS 4)

Author(s):  
Mir Ragib Ishraq ◽  
Nitesh Khadka ◽  
Asif Mohammed Samir ◽  
M. Shahidur Rahman

Three different Indic/Indo-Aryan languages - Bengali, Hindi and Nepali have been explored here in character level to find out similarities and dissimilarities. Having shared the same root, the Sanskrit, Indic languages bear common characteristics. That is why computer and language scientists can take the opportunity to develop common Natural Language Processing (NLP) techniques or algorithms. Bearing the concept in mind, we compare and analyze these three languages character by character. As an application of the hypothesis, we also developed a uniform sorting algorithm in two steps, first for the Bengali and Nepali languages only and then extended it for Hindi in the second step. Our thorough investigation with more than 30,000 words from each language suggests that, the algorithm maintains total accuracy as set by the local language authorities of the respective languages and good efficiency.


Author(s):  
Ewa Ropelewska ◽  
Wioletta Popińska ◽  
Kadir Sabanci ◽  
Muhammet Fatih Aslan

AbstractThe aim of this study was to build the discriminative models for distinguishing the different cultivars of flesh of pumpkin ‘Bambino’, ‘Butternut’, ‘Uchiki Kuri’ and ‘Orange’ based on selected textures of the outer surface of images of cubes. The novelty of research involved the use of about 2000 different textures for one image. The highest total accuracy (98%) of discrimination of pumpkin ‘Bambino’, ‘Butternut’, ‘Uchiki Kuri’ and ‘Orange’ was determined for models built based on textures selected from the color space Lab and the IBk classifier and some of the individual cultivars were classified with the correctness of 100%. The total accuracy of up to 96% was observed for color space RGB and 97.5% for color space XYZ. In the case of color channels, the total accuracies reached 91% for channel b, 89.5% for channel X, 89% for channel Z.


Author(s):  
I. Allaouzi ◽  
B. Benamrou ◽  
A. Allaouzi ◽  
M. Ouardouz ◽  
M. Ben Ahmed

Abstract. With the continued growth of confirmed cases of COVID-19, a highly infectious disease caused by a newly discovered coronavirus called Severe Acute Respiratory Syndrome Coronavirus 2, or SARS-CoV-2, there is an urgent need to find ways to help clinicians fight the virus by reducing the workload and speeding up the diagnosis of COVID-19. In this work, we propose an artificial intelligence solution “AI_COVID” which can help radiologists to know if the lungs are infected with the virus in just a few seconds.AI_COVID is based on a pre-trained DenseNet-121 model that detects subtle changes in the lungs and an SVM classifier that decides whether these changes are caused by COVID-19 or other diseases. AI_COVID is trained on thousands of frontal chest x-rays of people who have contracted COVID-19, healthy people, and people with viral or bacterial pneumonia. The experimental study is tested on 781 chest x-rays from two publicly available chest x-ray datasets COVID-19 radiography database and COVIDx Dataset. The performance results showed that our proposed model (DenseNet-121 + SVM) demonstrated high performance and yielded excellent results compared to the current methods in the literature, with a total accuracy of 99.74% and 98.85% for binary classification (COVID-19 vs. No COVID-19) and multi-class classification (COVID-19 vs. Normal vs. Pneumonia), respectively.


Author(s):  
Masar Abed Uthaib ◽  
Muayad Sadik Croock

In the classification of license plate there are some challenges such that the different sizes of plate numbers, the plates' background, and the number of the dataset of the plates. In this paper, a multiclass classification model established using deep convolutional neural network (CNN) to classify the license plate for three countries (Armenia, Belarus, Hungary) with the dataset of 600 images as 200 images for each class (160 for training and 40 for validation sets). Because of the small numbers of datasets, a preprocessing on the dataset is performed using pixel normalization and image data augmentation techniques (rotation, horizontal flip, zoom range) to increase the number of datasets. After that, we feed the augmented images into the convolution layer model, which consists of four blocks of convolution layer. For calculating and optimizing the efficiency of the classification model, a categorical cross-entropy and Adam optimizer used with a learning rate was 0.0001. The model's performance showed 99.17% and 97.50% of the training and validation sets accuracies sequentially, with total accuracy of classification is 96.66%. The time of training is lasting for 12 minutes. An anaconda python 3.7 and Keras Tensor flow backend are used.


Author(s):  
Harold Erbin ◽  
Riccardo Finotello ◽  
Robin Schneider ◽  
Mohamed Tamaazousti

Abstract We continue earlier efforts in computing the dimensions of tangent space cohomologies of Calabi-Yau manifolds using deep learning. In this paper, we consider the dataset of all Calabi-Yau four-folds constructed as complete intersections in products of projective spaces. Employing neural networks inspired by state-of-the-art computer vision architectures, we improve earlier benchmarks and demonstrate that all four non-trivial Hodge numbers can be learned at the same time using a multi-task architecture. With 30 % (80 %) training ratio, we reach an accuracy of 100 % for h(1,1) and 97 % for h(2,1) (100 % for both), 81 % (96 %) for h(3,1), and 49 % (83 %) for h(2,2). Assuming that the Euler number is known, as it is easy to compute, and taking into account the linear constraint arising from index computations, we get 100 % total accuracy.


2021 ◽  
Vol 910 (1) ◽  
pp. 012126
Author(s):  
Muzahim Saeed Younis ◽  
Saifaldeen Maadh Mustafa

Abstract The location indication was classified as a directed classification for all 44 species of trees present at the study site based on the projection of the tree coordinates on the false-color satellite images, which were taken from the location of these trees and their reflectivity measured in the laboratory. Where the satellite image was classified, based on the points taken for trees as field training areas, the visual output image classified by the directed classification method included 23 classes and represents the distribution of trees and shrubs at the site. The classification accuracy of vegetation and non-vegetation covers was also assessed by taking (334) ground control points for the various land targets and vegetation covers to determine this accuracy. Thus, we obtained a total accuracy of the classified statement (82.1%). This indicates that the accuracy of the overall classification is good, acceptable, and reliable. The percentage was high for all varieties, reaching (93%) for the frothy mug, weeping willow, and wild amethyst, and the lowest (75%) for olives, grassy slopes, and flat barren soils, and this was also acceptable. Through this accuracy, we can determine the extent to which the classification matches these goals and covers, and the possibility of relying on the prepared map for its future use. The number of each tree type was estimated by determining the coverage area for each tree type and the total area to cover the total type in the area using the proportional method. From this, it was found that the different types of trees differ in their presence on the site and the reason is attributed to the difference in height, direction and the different organic matter in which these types were grown and the environmental conditions appropriate to the species and that these factors have an effective role in the distribution of species and their densities in the different sites of the study area. We also noticed that the highest presence in terms of number was of edible oaks, followed by tannins oak in second place, at a rate ranging from (29.84%, 6.35%).


2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Gadissa Nemomsa ◽  
M. Azath

Nowadays, the huge amount of patient’s data significantly increases with respect to the time in repositories and data mining is increasingly used as an emerging research area in medical fields for extracting useful and previously unknown insights/patterns from the repository data. These unknown patterns/hidden insights can help in discovering new knowledge hidden in these data repositories. From the observation, different ARV regimens were ordered for different patients. However, combination of these drugs causes different side effects on the patients. It has been observed that there was a lack of predictive studies and designed models available in hospitals specifically ART Centers that accurately determine or classify the patient’s ARV regimen to TDF + 3TC + EFV, TDF + 3TC + NVP, AZT + 3TC + ATV/R, AZT + 3TC + LPV/R, TDF + 3TC + LVP/R, TDF + 3TC + ATV/R, 8888, and ABC + 3TC + LPV/R. In order to solve these kinds of problems, we built an accurate classifier system or model using parameters like Patient Age, Patient Encounter Day, Patient Encounter Month, Patient Encounter Year, Patient Weight, Patient CD4 Count Adult, Patient TB Screen, Patient Following WHO Stage, Patient CD4 Percent Child, Patient Regimen Specify, Patient Regimen, and so on. The general objective of this research was predictive modeling for the patient’s ARV regimen class through data mining techniques so as to improve them. The study used the CRIPS-DM methodology to find and interpret patterns in repositories. A decision tree (J48 and Random Forest) algorithm was used for classification. Using all tested classifiers, the investigation of the study shows that the total accuracy was more than 60%. On the other hand, among different classifications, class H (ABC + 3TC + LPV/R) has shown the worst prediction. But it was revealed that the J48 classifier relatively produces higher classification accuracy for the D (AZT-3TC-NVP) regimen. Here, classification depended on the selected parameters, which revealed that prediction accuracy value differed among all classifiers and the selected attributes. Finally, the study concluded that data mining can be used as a significant technique to discover patient regimen based on salient affecting factors with 96.1% precision achieved. Ensemble learning resolves the categorizing models of greater anticipating performance with different learning algorithms. This model aligned with sentimental investigation to magnify the appearances of the dataset either from the social media or from primary data collection. The empirical investigation with different parameters shows the detailed improvement of their learning methods.


2021 ◽  
Vol 10 (19) ◽  
pp. 4508
Author(s):  
Yoshitaka Kise ◽  
Chiaki Kuwada ◽  
Yoshiko Ariji ◽  
Munetaka Naitoh ◽  
Eiichiro Ariji

This study was performed to evaluate the diagnostic performance of deep learning systems using ultrasonography (USG) images of the submandibular glands (SMGs) in three different conditions: obstructive sialoadenitis, Sjögren’s syndrome (SjS), and normal glands. Fifty USG images with a confirmed diagnosis of obstructive sialoadenitis, 50 USG images with a confirmed diagnosis of SjS, and 50 USG images with no SMG abnormalities were included in the study. The training group comprised 40 obstructive sialoadenitis images, 40 SjS images, and 40 control images, and the test group comprised 10 obstructive sialoadenitis images, 10 SjS images, and 10 control images for deep learning analysis. The performance of the deep learning system was calculated and compared between two experienced radiologists. The sensitivity of the deep learning system in the obstructive sialoadenitis group, SjS group, and control group was 55.0%, 83.0%, and 73.0%, respectively, and the total accuracy was 70.3%. The sensitivity of the two radiologists was 64.0%, 72.0%, and 86.0%, respectively, and the total accuracy was 74.0%. This study revealed that the deep learning system was more sensitive than experienced radiologists in diagnosing SjS in USG images of two case groups and a group of healthy subjects in inflammation of SMGs.


Author(s):  
Ahmed Eid Fahim Abdella ◽  
Khaled Ismail Elshafey ◽  
Mohammed Fouad Sherif ◽  
Hanan Ahmad Nagy

Abstract Background Nowadays, PET/CT plays a substantial role in the diagnosis of different types of tumor by its ability to provide combined functional and anatomic imaging in the same session. The purpose of this study is to evaluate the added value of PET/CT in staging and re-staging of primary malignant bone tumors. Results Out of the studied 40 patients, 7 patients were referred for primary staging of different types of histologically proven primary malignant bone tumors, their FDG-PET/CT studies yielded additional diagnostic information in 28.6% of them. Thirty three patients were referred either for assessment of treatment response or for follow-up to detect any viable lesions; FDG-PET/CT was more sensitive and specific than CT in follow-up and assessment of treatment response with PET/CT sensitivity 94.4%, specificity 86.7%, and total accuracy 90.9% and CT sensitivity 88.2%, specificity 81.2%, and total accuracy 84.8%. Conclusions PET/CT was an accurate imaging modality in evaluation of primary malignant bone tumors regarding tumor staging, assessment of therapeutic response and detection of metastatic disease as compared to CT.


Author(s):  
Luigi Fassio ◽  
Longyang Lin ◽  
Raffaele De Rose ◽  
Marco Lanuzza ◽  
Felice Crupi ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document