Bilgisayar Mühendisliği Bölümü Koleksiyonu
Bu koleksiyon için kalıcı URI
Güncel Gönderiler
Öğe Nanonetwork-based search and rescue operations in debris areas(Elsevier, 2023) Şahin, Emre; Akkaş, Mustafa Alper; Dagdeviren, OrhanEarthquake is one of the most destructive natural disasters. It is vital to carry out search and rescue (SAR) operations in the debris area after such a major disaster. Although there are various technologies to rescue living things in the debris area, reaching narrow gaps in the rubble is a very difficult process. Nanonetworks operating on terahertz (THz) communication channels are composed of thousands of nanodevices that can pass through even ultrasmall gaps. In this paper, we propose an application that utilizes nanodevices for SAR operations in debris areas. To the best of our knowledge, our paper is the first study to provide a detailed and specialized analysis of this application. We work on channel models by extensively analyzing transmittance, propagation loss, absorption loss, path loss, signal-to-noise ratio (SNR), and capacity. Besides, we investigate a physical layer (PHY) model by considering the effects of using directional or omnidirectional antennas to ensure optimal throughput and energy consumption. Moreover, we present an exhaustive performance analysis of the modulation technique with respect to various nanodevice counts and densities.Öğe Classification of freezing of gait in Parkinson's disease using machine learning algorithms(Institute of Electrical and Electronics Engineers Inc., 2023) Önder, Mithat; Şentürk, Ümit; Polat, Kemal; Paulraj, D.Freezing of gait (FoG) is a prevalent and incapacitating symptom that affects individuals diagnosed with Parkinson's disease (PD) and other movement disorders. Detecting FoG is crucial for accurate diagnosis, fall prevention, and providing objective measurements, all of which are essential for optimizing treatment strategies and improving the quality of life for individuals with FoG. In this study, FoG has been detected using three different classification algorithms: Medium Gaussian Support Vector Machine (SVM), Medium K-Nearest Neighbor (KNN), and Boosted Trees. The process starts with data segmentation, where the dataset is divided into smaller segments. Then, feature extraction is performed on each segment to obtain various statistical measures such as mean, root mean square, maximum, standard deviation, kurtosis, skewness, and peak of root mean square. To ensure a robust and reliable analysis, the dataset is resampled using bootstrapping, a statistical technique that involves drawing random samples from the dataset with replacement. This leads to a more representative sample and reduces the impact of outliers or imbalanced data. The next step is to split the resampled dataset into three different approaches for the classification algorithm: In 5-FCV, the dataset is divided into five equal-sized subsets. Similarly, 10-FCV splits the dataset into ten subsets and follows the same process. Finally, the Medium Gaussian SVM, Medium KNN, and Boosted Trees classification algorithms are applied to the FoG dataset. The classification accuracy achieved is 86.9%, 87.6%, and 92.7% with 10-fold cross-validation, indicating that these algorithms are effective in accurately classifying FoG. © 2023 IEEE.Öğe Tip-I nöronlarda vibrasyonel rezonans vibrational resonance in type-I neurons(Institute of Electrical and Electronics Engineers Inc., 2022) Çalım, AliIn this study, vibrational resonance is discussed in the neural system with type-1 excitability. Vibrational resonance is defined as the nonlinear response of an oscillator system to a low frequency subthreshold weak signal with the help of a high frequency signal. Here, at an intermediate amplitude level of the high-frequency signal, the frequency of the oscillations is locked to the weak signal frequency, or the oscillation-signal coherence becomes maximum. In our work, we used Morris-Lecar neuron model as type-I excitable system and global network structure for connectivitiy. As a result of the simulations, it has been seen that in globally connected neural network, signal detection performance can increase more than three times compared to the performance of a single neuron. Different scenarios in synaptic conductivity and the application of high frequency signal are considered. Accordingly, these two parameters play a very active role in the detection and especially the propagation of weak signal or information in the network by means of information encoding of the neuron population based on vibrational resonance. © 2022 IEEE.Öğe Diagnosis of Alzheimer's disease using boosting classification algorithms(Institute of Electrical and Electronics Engineers Inc., 2023) Önder, Mithat; Şentürk, Ümit; Polat, Kemal; Paulraj, D.Alzheimer's Disease (AD) is a progressive degenerative disorder of the brain that impacts memory, cognition, and, ultimately, the ability to carry out daily activities. There is presently no cure for Alzheimer's Disease. However, there are available treatments to manage symptoms and slow their advancement. This research conducted a comprehensive study to diagnose AD using four different categorization methods. These methods included XGBoost, GradientBoost, AdaBoost, and voting classification algorithms. To carry out the examination, a high-quality dataset was obtained from the collection of machine learning data of the prestigious University of California. This dataset was carefully selected to ensure accurate and reliable results. The analysis of the collected data revealed some interesting findings. XGBoost exhibited an accuracy rate of 85% in diagnosing Alzheimer's Disease. ADABoost also performed, achieving an accuracy rate of 75%. GradientBoost, similarly, obtained an accuracy rate of 85%. Additionally, the voting classification algorithms showed promise, attaining an accuracy rate of 80%. All these accuracy rates were obtained by implementing a 5-fold cross-validation methodology, which ensured robust and unbiased results. This research contributes to the field of AD diagnosis by providing insights into the effectiveness of different categorization methods. © 2023 IEEE.Öğe Electricity consumption forecast using machine learning regression models in Turkey(Institute of Electrical and Electronics Engineers Inc., 2022) Şentürk, Ümit; Beken, Murat; Eyecioğlu, ÖnderIn today's energy crisis, countries need to know their energy consumption and make their energy investments accordingly. The variability of end users demanding energy makes it difficult to estimate energy needs. In this article, it has been tried to forecast the future consumption from the electrical energy data consumed in Turkey between the years 2016-2022. After the electricity consumption data was converted into daily data, electrical energy consumption estimations were made with machine learning methods such as linear regression, tree regression, voting regression, XGB regression and Artificial neural network (ANN) methods. Estimation results were evaluated with Mean Square Error (MSE) and R2 (coefficient of determination) performance metrics. As a result of the evaluations made with the test data, MSE=0.006 (0-1 min-max normalization dataset) and R2= 82.7 performances, voting regression obtained the best result among the methods used. Accurate estimation of energy consumption will enable energy production to be made at the optimum level. © 2022 IEEE.Öğe Investigating the energy production trends of countries and its relationship between economic development(Institute of Electrical and Electronics Engineers Inc., 2022) Hangün, Batuhan; Eyecioğlu, Önder; Beken, MuratThere are three main energy resources for energy production that is currently used by the countries can be listed as fossil fuels, renewable energy, and nuclear energy. Since the industrial revolution, the fundamental energy resource of the world is fossil sources. Due to their problems such as greenhouse gas emissions, production cost, etc., fossil fuels are not sustainable sources in terms of world ecology and economy. Developed countries of the world focused on using alternative energy resources like renewable energy and nuclear energy to produce green and sustainable energy for a better future. An assumption might be made about the relationship between the types of energy resources used by a country and its development level. In this study, we investigate the connection between the economic development of the countries and energy consumption patterns using a machine learning approach. We used a custom data set created by the combination of two separate data sets to deduce that relation by using the clustering method. It can be seen that developed countries are prone to cease the usage of fossil fuels and increase the energy produced by renewable and nuclear energy while underdeveloped or developing countries still rely mostly on fossil fuels. This trend among the developed countries has increased by the year. Additionally, we also wanted to see where Turkey stands among the selected countries and make an observation on its energy and economic development progress. © 2022 IEEE.Öğe Estimating the effect of renewable energy policies on the Republic of Turkey's gross national product by using artificial Intelligence(Institute of Electrical and Electronics Engineers Inc., 2022) Beken, Murat; Kurt, Nursaç; Eyecioğlu, ÖnderToday, most of the states in the energy field, especially the developed countries in the world, continue to increase the rate of use of renewable energy. The Republic of Turkey is one of these states. This study aimed to examine the effect of the renewable energy transformation process on the gross national product with artificial intelligence in the action plan of the Republic of Turkey in the process of moving away from fossil fuel-based energy sources. We deployed artificial neural networks to estimate Turkey's gross national product. As a result of the study, because of the policy of increasing the use of renewable energy sources by 7% annually and reducing the use of fossil derivative fuels by 8% annually in the next twenty years, the energy distributions in 2043 and their effect on Turkey's gross national product are shown. © 2022 IEEE.Öğe Beat estimation from musician visual cues(Sound and Music Computing Network, 2021) Chakraborty, Sutirtha; Aktaş, Senem; Clifford, William; Timoney, JosephMusical performance is an expressive art form where musicians interact with each other using auditory and nonverbal information. This paper aims to discover a robust technique that can identify musical phases (beats) through visual cues derived from a musician’s body movements captured through camera sensors.A multi-instrumental dataset was used to carry out a comparative study of two different approaches: (a) motiongram, and (b) pose-estimation, to detect phase from body sway. Decomposition and filtering algorithms were used to clean and fuse multiple signals. The final representations were analysed from which estimates of the beat, based on a’trust factor’, were obtained. The Motiongram and pose estimation were found to demonstrate usefulness depending on the musical instrument as some instrument playing gestures stimulate more movement in the players than others. Overall, the results were most promising using motiongram. It performed well where string instruments were used. The spatial derivative technique based on human pose estimation was consistent with woodwind instruments, where only a small degree of motion was observed. Copyright: © 2021 the Authors.Öğe Quantum computing approach to smart grid stability forecasting(Institute of Electrical and Electronics Engineers Inc., 2024) Hangün, Batuhan; Eyecioğlu, Önder; Altun, OğuzThe stability of a smart grid architecture is one of the most important parameters to assess its effectiveness and reliability. As a result of increased industrialization and use of renewable energy, predicting stability with respect to different scenarios becomes increasingly important to mitigate potential instabilities. Therefore, employing machine learning approaches to forecast system stability is crucial for sustainability. Traditionally, classical computers have been employed for machine learning-based methods; however, the proliferation of producers, consumers, and prosumers in newly built smart grid structures generates massive volumes of data, posing significant challenges for traditional computational methods in processing and analyzing these data efficiently. Quantum computing offers a promising solution for handling large amounts of data generated by smart grids, potentially leading to more accurate and efficient stability predictions. This study investigates a benchmark-based approach by comparing classical machine learning with a quantum machine learning technique to predict the stability of the smart grid. We utilized the "Electrical Grid Stability Simulated Data"dataset, focusing on a 4-node architecture smart grid. We approached this task as a classification problem, comparing the performance of a classical Support Vector Machine (SVM) model with a quantum machine learning approach, namely the Variational Quantum Classifier (VQC). Both methods were tested and evaluated on the basis of their ability to classify the grid's stability into "stable"and "unstable"categories. Our experimental results demonstrate the potential advantages and limitations of quantum computing in improving smart grid stability forecasting, and energy research in general. © 2024 IEEE.Öğe Stochastic resonance in a single autapse-coupled neuron(Pergamon-Elsevier Science Ltd, 2023) Baysal, Veli; Çalım, AliThe signal detection ability of nervous system is highly associated with nonlinear and collective behaviors in neuronal medium. Neuronal noise, which occurs as natural endogenous fluctuations in brain activity, is the most salient factor influencing this ability. Experimental and theoretical research suggests that noise is beneficial, not detrimental, for regular functioning of nervous system. In this regard, there is a general agreement that noise at an adequate intensity can engage rhythmic activity in brain and noise-induced oscillations enhances performance of the weak signal processing, especially when frequency of the signal is around that of the noise-induced rhythmic oscillation. This behavior in biological neural systems is explained by the notion of stochastic resonance. Another factor that plays a key role in regulating neuronal behaviors, including motor and cognitive tasks by maintaining signaling between cells, is characteristics of synapses different in structure and functioning. Here, we study stochastic resonance in Hodgkin-Huxley neuron that has a peculiar synaptic connection called autapse, known as a biophysical feedback mechanism, under presynaptic noise originating from superposition of inhibitory and excitatory Poisson bombardment. Our results show that, under certain conditions, autapse dynamics are able to improve the weak signal detection performance of Hodgkin-Huxley neuron via stochastic resonance. This study provides novel insights into functional role of autapse in neural information processing by revealing a biophysical aspect of stochastic resonance with numerical computations.Öğe Chaotic resonance in an astrocyte-coupled excitable neuron(Pergamon-Elseiver Science Ltd, 2023) Çalım, Ali; Baysal, VeliWe study the chaotic resonance phenomenon whereby the response of a neuron to a weak signal is amplified with the help of chaotic current stemming from background activity in the brain. This resonance behavior exhibits a bell-shaped curve in terms of detection quality due to increasing chaotic current intensity. Recent experimental studies have shown that astrocytes, which are the most abundant types of glial cells, may be responsible for the regulation of electrophysiological events in neuronal medium. Hence, we consider here a realistic neuronal system which is constituted by a bipartite network consisting of an excitable neuron and an astrocyte. Our analysis reveals that signal detection quality can be greatly enhanced with the astrocyte contribution obtained by appropriate neuronal and astrocytic cell dynamics. We find that depolarization -induced astrocytic glutamate release is able to improve chaotic resonance performance considerably in the presence of an adequately strong interaction between the astrocyte and the excitable neuron receiving a weak signal with a relatively higher frequency. We also show that a moderate production rate of gliotransmitters is required for the astrocyte to affect resonance performance of the neuron. Except for those conditions where the facilitating effect of astrocyte is observed, it can also reduce signal detection performance in the neuron. Furthermore, we demonstrated that intrinsic neuronal excitability is regulated by the astrocyte, via a comparison of resonance behaviors under effects of bias and astrocytic current separately. Taken together, our findings provide a novel insight into the functioning of astrocyte-neuron circuits, in particular the encoding weak signals via chaotic resonance, and suggest that astrocytes play a key role in intrinsic regulation and selectivity in neuronal information processing.Öğe Strain energy prediction of single wall carbon nanotubes using general regression neural network and adaptive neuro - Fuzzy Inference System(Taylor & Francis Inc, 2023) Eyecioğlu, Önder; Kayışlı, Korhan; Beken, MuratSingle Wall Carbon Nanotubes (SWCNTs) play crucial roles in the field of nanotechnology research and applications. Due to their quantum mechanical nature and the intricate structure of SWCNTs, performing direct or indirect experimental processes can be exceedingly challenging. Simulation methods like Tight-Binding Molecular Dynamics (TBMD) can serve as viable alternatives to experimentation. However, it's worth noting that these methods often demand extensive computational runtime. To address this computational time challenge, artificial intelligence algorithms such as the General Regression Neural Network (GRNN) and the Adaptive Neuro Fuzzy Interface System (ANFIS) have been proposed in this study. These models aim to calculate the energetic properties of SWCNTs more efficiently, offering practical and quicker predictive methods with reduced computational workloads. The study's findings demonstrate a strong correlation between the predicted energy values of SWCNTs using GRNN and ANFIS models and the results obtained through TBMD simulations. Consequently, it is believed that these models can be suitable and effective approaches for computing the energetic properties of SWCNTs.Öğe A study on interactions of 14.7-MeV protons and 3.6-MeV alphas in 93Nb target(Taylor & Francis Ltd, 2024) Yiğit, Mervenur; Kara, Ayhan; Yılmaz, AliNiobium is an important alloying material in nuclear reactors because of its enormous strength, low density, low neutron absorption, and high melting point. This study is structured on nuclear data calculations that are based on a Monte Carlo simulation approach. The GEANT4, SRIM, and TALYS codes were used to create a comprehensive simulation of 3.6-MeV alphas and 14.7-MeV protons on a Nb-93 target. We present calculation results on nuclear parameters as ion energy losses, displacements, vacancies, projected ranges, and cross sections. A comparison between the GEANT4 and SRIM codes was made for the projected ranges and ion energy losses. Besides, the calculations of cross sections in the TALYS code were carried out using level densities on the Skyrme energy density functional and the Fermi gas model.Öğe Analyzing the web sites of metropolitan municipalities in Turkey in terms of digital access indicators(Institute for Local Self-Government Maribor, 2023) Tutar, Hasan; Parlak, İsmail Hakkı; Tuzcuoğlu, Ferruh; Başpınar, Nuran Öztürk; Nam, SelçukE-government and e-municipality practices, which have gained critical importance in terms of the spread of citizen-oriented service and participatory management approach in local governments, have become a state policy in many countries. In Turkey, within the framework of restructuring the public administration and the e -transformation strategy, e-municipality practices are being expanded in local governments. One of the critical developments in the field of public administration in Turkey recently is e-government applications and e-municipality, which is its reflection on local governments. E -municipality applications are used as a local government strategy to quickly meet the citizens' municipal needs and reduce service costs. In this study, e-municipality services of metropolitan municipalities in Turkey; It is intended to be analyzed in terms of accessibility, performance, Searches Engine Optimization, and Best Practice indicators. The research findings show that the e-municipal services of metropolitan municipalities in Turkey should be improved significantly, given the mentioned indicators.Öğe SenDemonNet: Sentiment analysis for demonetization tweets using heuristic deep neural network(Springer, 2022) Kayıkçı, ŞafakSentiment analysis is one of the efficient models for extracting opinion mining with identification and classification from unstructured text data such as product reviews or microblogs. It is used to gain feedback from political campaigns, brand reviews, marketing analysis, and customers. The sentiment analysis on Twitter data is a recent research field in the natural processing. The dataset is gathered from the Twitter package in R along with Twitter API. The main intent of this paper is to understand the public opinion on the recently implemented demonetization policy using the proposed SenDemonNet. Initially, the tweet preprocessing was done, which is intended for cleaning the text data. Then, the feature extraction is performed by Bag of n-grams, TF-IDF, and the word2vec algorithm. The main objective of this work is a weighted feature selection that is developed by the hybrid Forest-Whale Optimization Algorithm (F-WOA) to get the best classification outcome. With these features, the Heuristic Deep Neural Network (HDNN) is adopted for classification, where the proposed FOA and WOA tune the parameter of DNN for reaching the maximum accuracy rate. From the statistical analysis, the performance of the designed F-WOA-DNN is 1.8%, 1.9%, 1.86%, and 2% enhanced than PSO-DNN, GWO-DNN, WOA-DNN, FOA-DNN, SVM, CNN, LSTM, and DNN respectively. Extensive experimental results show that SenDemonNet outperforms its competitors, producing an impressive increase in the classification accuracy on the benchmark dataset.Öğe Computational insights of bioconvective third grade nanofluid flow past a riga plate with triple stratification and swimming microorganisms(Hindawi Ltd, 2022) Kayıkçı, ŞafakThe goal of this study is to examine the heat-mass effects of a third grade nanofluid flow through a triply stratified medium containing nanoparticles and gyrostatic microorganisms swimming in the flow. The heat and mass fluxes are considered as a non-Fourier model. The governing models are constructed as a partial differential system. Using correct transformations, these systems are converted to an ordinary differential model. Ordinary systems are solved using convergent series solutions. The effects of physical parameters for fluid velocity, fluid temperature, nanoparticle volume percentage, motile microbe density, skin friction coefficients, local Nusselt number, and local Sherwood number are all illustrated in detail. When the values of the bioconvection Lewis number increase, the entropy rate also rises. The porosity parameter and modified Hartmann number show the opposite behaviour in the velocity profile.Öğe Diagnosis and classification of Parkinson's disease using ensemble learning and 1D-PDCovNN(Pergamon-Elsevier Science Ltd, 2023) Nour, Majid; Şentürk, Ümit; Polat, KemalIn this paper, we proposed a novel approach to diagnose and classify Parkinson's Disease (PD) using ensemble learning and 1D-PDCovNN, a novel deep learning technique. PD is a neurodegenerative disorder; early detection and correct classification are essential for better disease management. The primary aim of this study is to develop a robust approach to diagnosing and classifying PD using EEG signals. As the dataset, we have used the San Diego Resting State EEG dataset to evaluate our proposed method. The proposed method mainly consists of three stages. In the first stage, the Independent Component Analysis (ICA) method has been used as the pre-processing method to filter out the blink noises from the EEG signals. Also, the effect of the band showing motor cortex activity in the 7-30 Hz frequency band of EEG signals in diagnosing and classifying Parkinson's disease from EEG signals has been investigated. In the second stage, the Common Spatial Pattern (CSP) method has been used as the feature extraction to extract useful information from EEG signals. Finally, an ensemble learning approach, Dynamic Classifier Selection (DCS) in Modified Local Accuracy (MLA), has been employed in the third stage, consisting of seven different classifiers. As the classifier method, DCS in MLA, XGBoost, and 1D-PDCovNN classifier has been used to classify the EEG signals as the PD and healthy control (HC). We first used dynamic classifier selection to diagnose and classify Parkinson's disease (PD) from EEG signals, and promising results have been obtained. The performance of the proposed approach has been evaluated using the classification accuracy, F-1 score, kappa score, Jaccard score, ROC curve, recall, and precision values in the classification of PD with the proposed models. In the classification of PD, the combination of DCS in MLA achieved an accuracy of 99,31%. The results of this study demonstrate that the proposed approach can be used as a reliable tool for early diagnosis and classification of PD.Öğe A novel cuffless blood pressure prediction: Uncovering new features and new hybrid ml models(MDPI, 2023) Nour, Majid; Polat, Kemal; Şentürk, Ümit; Arıcan, MuratThis paper investigates new feature extraction and regression methods for predicting cuffless blood pressure from PPG signals. Cuffless blood pressure is a technology that measures blood pressure without needing a cuff. This technology can be used in various medical applications, including home health monitoring, clinical uses, and portable devices. The new feature extraction method involves extracting meaningful features (time and chaotic features) from the PPG signals in the prediction of systolic blood pressure (SBP) and diastolic blood pressure (DBP) values. These extracted features are then used as inputs to regression models, which are used to predict cuffless blood pressure. The regression model performances were evaluated using root mean squared error (RMSE), R-2, mean square error (MSE), and the mean absolute error (MAE). The obtained RMSE was 4.277 for systolic blood pressure (SBP) values using the Matern 5/2 Gaussian process regression model. The obtained RMSE was 2.303 for diastolic blood pressure (DBP) values using the rational quadratic Gaussian process regression model. The results of this study have shown that the proposed feature extraction and regression models can predict cuffless blood pressure with reasonable accuracy. This study provides a novel approach for predicting cuffless blood pressure and can be used to develop more accurate models in the future.Öğe Machine learning models for estimating the AC conductivity mechanism of Edirne Kufeki stone reinforced Polypyrrole composites(WILEY, 2022) Eyecioğlu, ÖnderIn this study, the detailed temperature and frequency-dependent (ac) conductivity analyzes of Polypyrrole/Edirne Kufeki Stone (PPy/EKS) composites have been realized by considering both experimental and Machine Learning (ML) algorithms predicted data. In this respect, the experimental ac conductivity data of pure PPy, PPy/5% EKS, PPy/10% EKS, and PPy/20% EKS composites between 1 Hz and 40 MHz at 296, 313, and 333 K temperatures have been used for the data set of ML. First, a benchmark study has been done for applied ML algorithms to obtain an eligible model. It is found that the Gaussian process regression (GPR) algorithm provided the best prediction performance. Since it has been observed a good conformity between the experimental and prediction data of GPR model, the ac conductivity (sigma(ac)) versus angular frequency (w) curves of the composites produced experimentally have been estimated for new temperature values, which were not treated experimentally. Then, the f(w) curves of at temperature values have been estimated by GPR which is for the EKS composites at various contributions that have not been experimentally produced. Ultimately, the GPR algorithm developed in the present work enables us to determine the optimum EKS additive percentage, working temperature, and frequency band for the PPy polymer matrix.Öğe Thermal analysis of radiative water- and glycerin-based carbon nanotubes past a Riga plate with stratification and non-Fourier heat flux theory(Springer, 2023) Kayıkçı, Şafak; Eswaramoorthi, S.; Postalcıoğlu, Seda; Loganathan, K.The impact of thermal stratification and thermally radiative flow of carbon nanotubes on a Riga plate with injection/suction and heat generation/consumption is investigated. The two varieties of base fluids, like, water and glycerin with single-wall nanotubes and multi-wall carbon nanotubes, are incorporated in this investigation. Cattaneo-Christov heat flux theory is utilized to frame the energy equation. The controlling PDEs are remodeled into ODEs using the appropriate variables. The obtained ODEs are analytically solved by applying the HAM procedure and numerically solved by using the BVP4c scheme. The consequences of the physical parameters on fluid velocity, fluid temperature, skin friction coefficients and local Nusselt number are explained through tables, graphs and charts. It is detected that the fluid velocity in both directions diminishes when raising the suction/injection and velocity slip parameters. The fluid temperature downturns when enhancing the suction/injection and stratification parameters. The surface shear stress suppresses when increasing the Forchheimer number. The radiation parameter leads to strengthening the heat transfer gradient.
