Bilgisayar Mühendisliği Bölümü Bildiri Koleksiyonu
Bu koleksiyon için kalıcı URI
Güncel Gönderiler
Öğe Classification of stroke using machine learning techniques : review study(IEEE, 2023) Sawan, Aktham; Awad, Mohammed; Qasrawi, RadwanAbstract—Presently, stroke is the leading cause of adult injury worldwide. The World Health Organization estimates that each year 15 million people around the world suffer a stroke. Five million of them die, and another five million are disabled for life. There is a chance to dramatically enhance the classification of strokes in the early stages. In this article, we reviewed all portable devices that produced electroencephalogram(EEG) data and all machine learning (ML) methods and deep-learning methods used to identify stroke using EEG data, and we noted that the amount of work on ML and deep learning in analyzing EEG data have increased rapidly in recent years. Such analysis has achieved greater precision compared to that conventional methods. We also discussed in this study the opportunities and key challenges for improving the accuracy of future work.Öğe A critical evaluation on nuclear safety properties of novel cadmium oxide-rich glass containers for transportation and waste management: benchmarking with a reinforced concrete container(FRONTIERS MEDIA SA, 2022) Almisned, Ghada; Şen Baykal, Duygu; Kılıç, Gökhan; İlik, Erkan; Zakaly, Hesham M. H.; Ene, Antoaneta; Tekin, Hüseyin OzanWe examine the nuclear safety properties of a newly designed cadmiumoxide-rich glass container for nuclear material to a bitumen-reinforced concrete container. Individual transmission factors, detectormodelling, and energy deposition (MeV/g) in the air are calculated using MCNPX (version 2.7.0) general purpose Monte Carlo code. Two container configurations are designed with the material properties of cadmium dioxide-rich glass and Concrete + Bitument in consideration. First, individual transmission factors for 60Co and 137Cs radioisotopes are calculated. To evaluate potential environmental consequences, energy deposition amounts in the air for 60Co and 137Cs are also determined. The minimum gamma-ray transmission rates for two container types are reported for a cadmium dioxiderich glass container. In addition, the quantity of energy deposition is varied depending on the container type, with a lower value for cadmium dioxide-rich glass container. The 40% cadmium dioxide-doped glass container provides more effective safety than the Cement + Bitumen container, according to the overall findings. In conclusion, the utilization of cadmium dioxide-doped glass material along with its high transparency and advanced material properties may be a significant and effective option in areas where concrete is required to assure the safety of nuclear materials.Öğe Gadolinium-tungsten-boron trioxide glasses: A multi-phase research on cross-sections, attenuation coefficients, build-up factors and individual transmission factors using MCNPX(Elsevier, 2022) ALMisned, Ghada; Sen Baykal, Duygu; Ali, Fatema T.; Bilal, Ghaida; Kılıç, Gökhan; Tekin, Hüseyin OzanThe oxide of the rare earth element gadolinium has the chemical formula Gd2O3. Also known as gadolinium sesquioxide, gadolinium trioxide, and Gadolinia, gadolinium oxide. In this study, various types of fundamental cross-sections, attenuation coefficients, build-up factors and individual transmission factors of high density gadolinium-tungsten-boron trioxide glasses with a chemical composition of (70-x)WO3-xGd2O3 –30B2O3 (where x: 17.5, 20.0, 22.5, 25.0 and 27.5 mol%) are determined using advanced Monte Carlo methods. In addition, gamma transmission factors (TFs) for a range of medical and industrial radioisotopes were calculated using MCNPX (version 2.7.0) Monte Carlo code. The investigated glasses were classified Gd17.5, Gd20.0, Gd22.5, Gd25.0, and Gd27.5 in accordance with xGd2O3. Our findings suggest that the Gd27.5 sample (with highest of Gd2O3 content mol. %) has possessed the maximum linear (µ) and mass (µ/?) attenuation coefficients at all gamma-ray energies investigated. The coded glass sample Gd27.5 is achieved the maximum effective atomic number (Zeff) and effective electron density (Neff) owing its superior attenuation properties. In terms of build-up factors, increasing the concentration of xGd2O3 in glasses is decreased the EBF and EABF values for all mean free path values (0.5–40 mfp). At a thickness of 3 cm, the lowest transmission factor (i.e., highest attenuation) was verified for all Gd17.5-Gd27.5 glasses investigated. Consequently, the Gd27.5 sample exhibits superior radiation shielding properties for a large range of photon energy and various medical and industrial radioisotope energies. © 2022 Elsevier GmbHÖğe Impact of COVID-19 lockdown on smoking (waterpipe and cigarette) and participants' BMI across various sociodemographic groups in Arab countries in the mediterranean region(2022) Al Sabbah, H.; Assaf, E. A.; Taha, Z.; Qasrawi, R.; İsmail, L. C.; Al Dhaheri, A. S.; Al-Mannai, M.INTRODUCTION Tobacco smokers are at high risk of developing severe COVID-19. Lockdown was a chosen strategy to deal with the spread of infectious diseases; nonetheless, it influenced people’s eating and smoking behaviors. The main objective of this study is to determine the impact of the COVID-19 lockdown on smoking (waterpipe and cigarette) behavior and its associations with sociodemographic characteristics and body mass index. METHODS The data were derived from a large-scale retrospective cross-sectional study using a validated online international survey from 38 countries (n=37207) conducted between 17 April and 25 June 2020. The Eastern Mediterranean Region (WHO-EMR countries) data related to 10 Arabic countries that participated in this survey have been selected for analysis in this study. A total of 12433 participants were included in the analysis of this study, reporting their smoking behavior and their BMI before and during the COVID-19 lockdown. Descriptive and regression analyses were conducted to examine the association between smoking practices and the participant’s country of origin, sociodemographic characteristics, and BMI (kg/m2 ). RESULTS Overall, the prevalence rate of smoking decreased significantly during the lockdown from 29.8% to 23.5% (p<0.05). The percentage of females who smoke was higher than males among the studied population. The highest smoking prevalence was found in Lebanon (33.2%), and the lowest was in Oman (7.9%). In Egypt, Kuwait, Lebanon, and Saudi Arabia, the data showed a significant difference in the education level of smokers before and during the lockdown (p<0.05). Smokers in Lebanon had lower education levels than those in other countries, where the majority of smokers had a Bachelor’s degree. The findings show that the BMI rates in Jordan, Lebanon, Oman, and Saudi Arabia significantly increased during the lockdown (p<0.05). The highest percentages of obesity among smokers before the lockdown were in Oman (33.3%), followed by Bahrain (28.4%) and Qatar (26.4%), whereas, during the lockdown, the percentage of obese smokers was highest in Bahrain (32.1%) followed by Qatar (31.3%) and Oman (25%). According to the logistic regression model, the odds ratio of smoking increased during the pandemic, whereas the odds ratio of TV watching decreased. This finding was statistically significant by age, gender, education level, country of residence, and work status. CONCLUSIONS Although the overall rates of smoking among the studied countries decreased during the lockdown period, we cannot attribute this change in smokingbehavior to the lockdown. Smoking cessation services need to anticipate that unexpected disruptions, such as pandemic lockdowns, may be associated with changes in daily tobacco consumption. Public health authorities should promote the adoption of healthy lifestyles to reduce the long-term negative effects of the lockdown.Öğe Algorithm selection for the team orienteering problem(SPRINGER-VERLAG BERLIN, 2022) Mısır, Mustafa; Gunawan, Aldy; Vansteenwegen, PieterThis work utilizes Algorithm Selection for solving the Team Orienteering Problem (TOP). The TOP is an NP-hard combinatorial optimization problem in the routing domain. This problem has been modelled with various extensions to address different real-world problems like tourist trip planning The complexity of the problem motivated to devise new algorithms. However, none of the existing algorithms came with the best performance across all the widely used benchmark instances. This fact suggests that there is a performance gap to fill. This gap can be targeted by developing more new algorithms as attempted by many researchers before. An alternative strategy is performing Algorithm Selection that will automatically choose the most appropriate algorithm for a given problem instance. This study considers the existing algorithms for the Team Orienteering Problem as the candidate method set. For matching the best algorithm with each problem instance, the specific instance characteristics are used as the instance features. An algorithm Selection approach, namely ALORS, is used to conduct the selection mission. The computational analysis based on 157 instances showed that Algorithm Selection outperforms the state-of-the-art algorithms despite the simplicity of the Algorithm Selection setting. Further analysis illustrates the match between certain algorithms and certain instances. Additional analysis showed that the time budget significantly affects the algorithms' performance.Öğe User experience and performance evaluation of palestinian universities websites(Institute of Electrical and Electronics Engineers Inc., 2021) Qasrawi, Aysar; Vicunapolo, Stephanny; Qasrawi, RadwanUser Experience and Performance analysis of websites are key factors in their accessibility and usability assessment. According to the human-computer interaction standards and guidelines, the user experience and performance can be affected by several factors such as page speed, web design, responsiveness, usability, accessibility, and security. Recently, the user experience and performance were conducted using automatic quality assurance tools. In this research paper, we aim to evaluate and compare the user experience and performance of Palestinian universities' websites by using the ImmuniWeb software, and JMeter software automated tools, as well as to validate the result through traditional survey user experience evaluation tool. The study evaluated response rates and security performance on a sample of 4 Palestinian university websites. The websites were also assessed in terms of efficiency, effectiveness, and user satisfaction through a comprehensive survey conducted on a sample of 84 students. The results indicated that the universities' websites had several challenges in terms of efficiency, effectiveness, and user satisfaction. Thus, Universities must upgrade their security features, use automation tools to tests their performance, and enhance their overall usability. © 2021 IEEE.Öğe Selection-based per-instance heuristic generation for protein structure prediction of 2D HP model(Institute of Electrical and Electronics Engineers Inc., 2021) Mısır, MustafaThe present study aims at generating heuristics for Protein Structure Prediction represented in the 2D HP model. Protein Structure Prediction is about determining the 3-dimensional form of a protein from a given amino acid sequence. The resulting structure directly relates to the functionalities of the protein. There are a wide range of algorithms to address Protein Structure Prediction as an optimization problem. Being said that there is no an ultimate algorithm that can effectively solve PSP under varying experimental settings. Hyper-heuristics can offer a solution as high-level, problem-independent search and optimization strategies. Selection Hyper-heuristics operate on given heuristic sets that directly work on the solution space. One group of Selection Hyper-heuristics focus on automatically specify the best heuristics on-the-fly. Yet, the candidate heuristics tend to be decided, preferably a domain expert. Generation Hyper-heuristics approach differently as aiming to generate such heuristics automatically. This work introduces a automated heuristic generation strategy supporting Selection Hyper-heuristics. The generation task is formulated as a selection problem, disclosing the best expected heuristic specifically f or a given problem instance. The heuristic generation process is established as a parameter configuration problem. T he corresponding system is devised by initially generating a training data alongside with a set of basic features characterizing the Protein Structure Prediction problem instances. The data is generated discretizing the parameter configuration space o f a single heuristic. The resulting data is used to predict the best configuration of a specific heuristic used in a heuristic set under Selection Hyper-heuristics. The prediction is performed separately for each instance rather than using one setting for all the instances. The empirical analysis showed that the proposed idea offers both better and robust performance on 22 PSP instances compared to the one-for-all heuristic sets. Additional analysis linked to the selection method, ALORS, revealed insights on what makes the PSP instances hard / easy while providing dis/-similarity analysis between the candidate configurations. © 2021 IEEE.Öğe Estimation of gas emission values on highways in Turkey with machine learning(IEEE, 2021) Kurt, Nursaç; Ozturk, O.; Beken, M.Due to its geographical location, Turkey has been home to many civilizations for centuries. It has always acted as a bridge between west and east and will continue to do so. The development of road networks in Turkey and the difference in transportation methods are increasing the number of national and international traveling vehicles day by day. In this study, gas emission (CO2, CH4, N2O) value changes have been predicted according to vehicle types of vehicle mobility on highways using machine learning (Linear Regression, Bayesian Ridge, Random Forest Regressor, MLP Regressor, SVR) algorithms. Based on these results, the gas emission value and environmental impact that may occur in the future are estimated - each method evaluated with MAE, MSE, RMSE, and R2 statistical metrics. As a result, we obtain R square scores of 0.963231 for CO2, 0.9856 for CH4, and 0.982404 for N2O from the random forest regressor, random forest regressor, and MLP regressor, respectively. © 2021 IEEE.Öğe Algorithm selection on adaptive operator selection : a case study on genetic algorithms(Springer Science and Business Media Deutschland GmbH, 2021) Mısır, MustafaThe present study applies Algorithm Selection (AS) to Adaptive Operator Selection (AOS) for further improving the performance of the AOS methods. AOS aims at delivering high performance in solving a given problem through combining the strengths of multiple operators. Although the AOS methods are expected to outperform running each operator separately, there is no one AOS method can consistently perform the best. Thus, there is still room for improvement which can be provided by using the best AOS method for each problem instance being solved. For this purpose, the AS problem on AOS is investigated. The underlying AOS methods are applied to choose the crossover operator for a Genetic Algorithm (GA). The Quadratic Assignment Problem (QAP) is used as the target problem domain. For carrying out AS, a suite of simple and easy-to-calculate features characterizing the QAP instances is introduced. The corresponding empirical analysis revealed that AS offers improved performance and robustness by utilizing the strenghts of different AOS approaches. © 2021, Springer Nature Switzerland AG.Öğe Active matrix completion for algorithm selection(Springer, 2019) Mısır, MustafaThe present work accommodates active matrix completion to generate cheap and informative incomplete algorithm selection datasets. Algorithm selection is being used to detect the best possible algorithm(s) for a given problem ((formula presented) instance). Although its success has been shown in varying problem domains, the performance of an algorithm selection technique heavily depends on the quality of the existing dataset. One critical and likely to be the most expensive part of an algorithm selection dataset is its performance data. Performance data involves the performance of a group of algorithms on a set of instance of a particular problem. Thus, matrix completion [1] has been studied to be able to perform algorithm selection when the performance data is incomplete. The focus of this study is to come up with a strategy to generate/sample low-cost, incomplete performance data that can lead to effective completion results. For this purpose, a number of matrix completion methods are utilized in the form of active matrix completion. The empirical analysis carried out on a set of algorithm selection datasets revealed significant gains in terms of the computation time, required to produce the relevant performance data. © Springer Nature Switzerland AG 2019.
