Preferred Language
Articles
/
ijs-8257
Attention Mechanism Based on a Pre-trained Model for Improving Arabic Fake News Predictions
...Show More Authors

     Social media and news agencies are major sources for tracking news and events. With these sources' massive amounts of data, it is easy to spread false or misleading information. Given the great dangers of fake news to societies, previous studies have given great attention to detecting it and limiting its impact. As such, this work aims to use modern deep learning techniques to detect Arabic fake news. In the proposed system, the attention model is adapted with bidirectional long-short-term memory (Bi-LSTM) to identify the most informative words in the sentence. Then, a multi-layer perceptron (MLP) is applied to classify news articles as fake or real. The experiments are conducted on a newly launched Arabic dataset called the Arabic Fake News Dataset (AFND). The AFDN dataset contains exactly 606912 news articles collected from multiple sources, so it is suitable for deep learning requirements. Both simple recurrent neural networks (S-RNN), long short-term memory (LSTM), and gated recurrent units (GRU) are used for comparison. According to evaluation criteria, our proposed model achieved an accuracy of (0.8127), which is the best and highest accuracy among the deep learning methods used in this work. Moreover, the performance of our proposed model is better compared to previous studies, which used the AFND.

Scopus Crossref
View Publication Preview PDF
Quick Preview PDF
Publication Date
Wed Mar 01 2017
Journal Name
Neural Computing And Applications
The potential of nonparametric model in foundation bearing capacity prediction
...Show More Authors

View Publication
Scopus (11)
Crossref (9)
Scopus Clarivate Crossref
Publication Date
Fri Aug 23 2013
Journal Name
International Journal Of Computer Applications
Lossless Compression of Medical Images using Multiresolution Polynomial Approximation Model
...Show More Authors

In this paper, a simple fast lossless image compression method is introduced for compressing medical images, it is based on integrates multiresolution coding along with polynomial approximation of linear based to decompose image signal followed by efficient coding. The test results indicate that the suggested method can lead to promising performance due to flexibility in overcoming the limitations or restrictions of the model order length and extra overhead information required compared to traditional predictive coding techniques.

View Publication
Crossref (4)
Crossref
Publication Date
Sat Oct 20 2018
Journal Name
Journal Of Economics And Administrative Sciences
Bayesian Tobit Quantile Regression Model Using Four Level Prior Distributions
...Show More Authors

Abstract:

      In this research we discussed the parameter estimation and variable selection in Tobit quantile regression model in present of multicollinearity problem. We used elastic net technique as an important technique for dealing with both multicollinearity and variable selection. Depending on the data we proposed Bayesian Tobit hierarchical model with four level prior distributions . We assumed both tuning parameter are random variable and estimated them with the other unknown parameter in the model .Simulation study was used for explain the efficiency of the proposed method and then we compared our approach with (Alhamzwi 2014 & standard QR) .The result illustrated that our approach

... Show More
View Publication Preview PDF
Crossref
Publication Date
Fri Dec 08 2023
Journal Name
Iraqi Journal Of Science
Molecules mixture model to explain the nature of interstellar matter
...Show More Authors

The nature of interstellar matter has been explained by presenting a model consists of organic and inorganic materials mixture in the wavelength range (9-12) μm.
Laboratory samples of different concentrations of these materials were prepared and examined by using FTIR spectrometer device. Results of spectra for these samples are compared with observation of Trapezium nebula in the same wavelengths range. The best mixture model found to fit with observation is a sample consists of:11.96 % (Diatom silica (98% pure silica)) + 14.35 % (Carbon (C)) +27.63 % (Tryptophan amino acid) + 46.06 % (Tyrosine amino acid), this are done using convolution technique. Interstellar matter could be explained biologically, as the contribution of organic

... Show More
View Publication Preview PDF
Publication Date
Sun Mar 01 2015
Journal Name
Journal Of Engineering
Bearing Capacity of Bored Pile Model Constructed in Gypseous Soil
...Show More Authors

Gypseous soils are distributed in many regions in the world including Iraq, which cover more than (31%) of the surface area of the country. Existence of these soils, always with high gypsum content, caused difficult problems to the buildings and strategic projects due to dissolution and leaching of the gypsum caused by the action of water flow through soil mass. For the study, the gypseous soil was brought from Bahr Al-Najaf, Al-Najaf Governorate which is located in the middle of Iraq. The model pile was embedded in gypseous soil with 42% gypsum content. Compression axial model pile load tests have been carried out for model pile embedded in gypseous soil at initial degree of saturation of (7%) before and after soil satu

... Show More
View Publication Preview PDF
Publication Date
Fri Sep 30 2022
Journal Name
Iraqi Journal Of Science
Optimal Control Design of the In-vivo HIV Fractional Model
...Show More Authors

    HIV is a leading cause of death, in particular, in Sub-Saharan Africa. In this paper, a fractional differential system in vivo deterministic models for HIV dynamics is presented and analyzed. The main roles played by different HIV treatment methods are investigated using fractional optimal control theory. We use three treatment regimens as system control variables to determine the best strategies for controlling the infection. The optimality system is numerically solved using the fractional Adams-Bashforth technique.

View Publication Preview PDF
Scopus (19)
Crossref (14)
Scopus Crossref
Publication Date
Fri Sep 30 2022
Journal Name
Journal Of Economics And Administrative Sciences
Robust Estimation OF The Partial Regression Model Using Wavelet Thresholding
...Show More Authors

            Semi-parametric regression models have been studied in a variety of applications and scientific fields due to their high flexibility in dealing with data that has problems, as they are characterized by the ease of interpretation of the parameter part while retaining the flexibility of the non-parametric part. The response variable or explanatory variables can have outliers, and the OLS approach have the sensitivity to outliers. To address this issue, robust (resistance) methods were used, which are less sensitive in the presence of outlier values in the data. This study aims to estimate the partial regression model using the robust estimation method with the wavel

... Show More
View Publication Preview PDF
Crossref
Publication Date
Sun Nov 01 2015
Journal Name
Journal Of Engineering
Greening Existing Buildings in Contemporary Iraqi Urban Reality/ Virtual Model
...Show More Authors

The approach of greening existing buildings, is an urgent necessity, because the greening operation provides the speed and optimal efficiency in the environmental performance, as well as keeping up with the global green architecture revolution. Therefore, greening existing buildings in Iraq is important for trends towards renewable energies, because of what the country went through economic conditions and crises and wars which kept the country away from what took place globally in this issue. The research problem is: insufficient knowledge about the importance and the mechanism of the greening of existing buildings, including its environmental and economic dimensions, by rationalization of energy consumption and preservi

... Show More
View Publication Preview PDF
Publication Date
Thu Apr 30 2020
Journal Name
Journal Of Economics And Administrative Sciences
Estimate the Partial Linear Model Using Wavelet and Kernel Smoothers
...Show More Authors

This article aims to estimate the partially linear model by using two methods, which are the Wavelet and Kernel Smoothers. Simulation experiments are used to study the small sample behavior depending on different functions, sample sizes, and variances. Results explained that the wavelet smoother is the best depending on the mean average squares error criterion for all cases that used.

 

 

View Publication Preview PDF
Crossref
Publication Date
Thu Sep 01 2022
Journal Name
Iaes International Journal Of Robotics And Automation
Implementation of a complex fractional order proportional-integral-derivative controller for a first order plus dead time system
...Show More Authors

This paper presents the implementation of a complex fractional order proportional integral derivative (CPID) and a real fractional order PID (RPID) controllers. The analysis and design of both controllers were carried out in a previous work done by the author, where the design specifications were classified into easy (case 1) and hard (case 2) design specifications. The main contribution of this paper is combining CRONE approximation and linear phase CRONE approximation to implement the CPID controller. The designed controllers-RPID and CPID-are implemented to control flowing water with low pressure circuit, which is a first order plus dead time system. Simulation results demonstrate that while the implemented RPID controller fails to stabi

... Show More