This research deals with a shrinking method concerned with the principal components similar to that one which used in the multiple regression “Least Absolute Shrinkage and Selection: LASS”. The goal here is to make an uncorrelated linear combinations from only a subset of explanatory variables that may have a multicollinearity problem instead taking the whole number say, (K) of them. This shrinkage will force some coefficients to equal zero, after making some restriction on them by some "tuning parameter" say, (t) which balances the bias and variance amount from side, and doesn't exceed the acceptable percent explained variance of these components. This had been shown by MSE criterion in the regression case and the percent explained variance in the principal component case.
Data centric techniques, like data aggregation via modified algorithm based on fuzzy clustering algorithm with voronoi diagram which is called modified Voronoi Fuzzy Clustering Algorithm (VFCA) is presented in this paper. In the modified algorithm, the sensed area divided into number of voronoi cells by applying voronoi diagram, these cells are clustered by a fuzzy C-means method (FCM) to reduce the transmission distance. Then an appropriate cluster head (CH) for each cluster is elected. Three parameters are used for this election process, the energy, distance between CH and its neighbor sensors and packet loss values. Furthermore, data aggregation is employed in each CH to reduce the amount of data transmission which le
... Show MoreIn this research we solved numerically Boltzmann transport equation in order to calculate the transport parameters, such as, drift velocity, W, D/? (ratio of diffusion coefficient to the mobility) and momentum transfer collision frequency ?m, for purpose of determination of magnetic drift velocity WM and magnetic deflection coefficient ? for low energy electrons, that moves in the electric field E, crossed with magnetic field B, i.e; E×B, in the nitrogen, Argon, Helium and it's gases mixtures as a function of: E/N (ratio of electric field strength to the number density of gas), E/P300 (ratio of electric field strength to the gas pressure) and D/? which covered a different ranges for E/P300 at temperatures 300°k (Kelvin). The results show
... Show MoreThis study aims to estimate the accuracy of digital elevation models (DEM) which are created with exploitation of open source Google Earth data and comparing with the widely available DEM datasets, Shuttle Radar Topography Mission (SRTM), version 3, and Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM), version 2. The GPS technique is used in this study to produce digital elevation raster with a high level of accuracy, as reference raster, compared to the DEM datasets. Baghdad University, Al Jadriya campus, is selected as a study area. Besides, 151 reference points were created within the study area to evaluate the results based on the values of RMS.Furthermore, th
... Show MoreModern civilization increasingly relies on sustainable and eco-friendly data centers as the core hubs of intelligent computing. However, these data centers, while vital, also face heightened vulnerability to hacking due to their role as the convergence points of numerous network connection nodes. Recognizing and addressing this vulnerability, particularly within the confines of green data centers, is a pressing concern. This paper proposes a novel approach to mitigate this threat by leveraging swarm intelligence techniques to detect prospective and hidden compromised devices within the data center environment. The core objective is to ensure sustainable intelligent computing through a colony strategy. The research primarily focusses on the
... Show MoreIn this paper, we introduce three robust fuzzy estimators of a location parameter based on Buckley’s approach, in the presence of outliers. These estimates were compared using the variance of fuzzy numbers criterion, all these estimates were best of Buckley’s estimate. of these, the fuzzy median was the best in the case of small and medium sample size, and in large sample size, the fuzzy trimmed mean was the best.
The Dagum Regression Model, introduced to address limitations in traditional econometric models, provides enhanced flexibility for analyzing data characterized by heavy tails and asymmetry, which is common in income and wealth distributions. This paper develops and applies the Dagum model, demonstrating its advantages over other distributions such as the Log-Normal and Gamma distributions. The model's parameters are estimated using Maximum Likelihood Estimation (MLE) and the Method of Moments (MoM). A simulation study evaluates both methods' performance across various sample sizes, showing that MoM tends to offer more robust and precise estimates, particularly in small samples. These findings provide valuable insights into the ana
... Show MoreIn this article, we developed a new loss function, as the simplification of linear exponential loss function (LINEX) by weighting LINEX function. We derive a scale parameter, reliability and the hazard functions in accordance with upper record values of the Lomax distribution (LD). To study a small sample behavior performance of the proposed loss function using a Monte Carlo simulation, we make a comparison among maximum likelihood estimator, Bayesian estimator by means of LINEX loss function and Bayesian estimator using square error loss (SE) function. The consequences have shown that a modified method is the finest for valuing a scale parameter, reliability and hazard functions.
The data preprocessing step is an important step in web usage mining because of the nature of log data, which are heterogeneous, unstructured, and noisy. Given the scalability and efficiency of algorithms in pattern discovery, a preprocessing step must be applied. In this study, the sequential methodologies utilized in the preprocessing of data from web server logs, with an emphasis on sub-phases, such as session identification, user identification, and data cleansing, are comprehensively evaluated and meticulously examined.
One of the costliest problems facing the production of hydrocarbons in unconsolidated sandstone reservoirs is the production of sand once hydrocarbon production starts. The sanding start prediction model is very important to decide on sand control in the future, including whether or when sand control should be used. This research developed an easy-to-use Computer program to determine the beginning of sanding sites in the driven area. The model is based on estimating the critical pressure drop that occurs when sand is onset to produced. The outcomes have been drawn as a function of the free sand production with the critical flow rates for reservoir pressure decline. The results show that the pressure drawdown required to
... Show MoreJournal of Theoretical and Applied Information Technology is a peer-reviewed electronic research papers & review papers journal with aim of promoting and publishing original high quality research dealing with theoretical and scientific aspects in all disciplines of IT (Informaiton Technology