DSpace Arşivi :: by Yazar "Uçan, Osman Nuri" değerine göre listeleniyor

Yazar "Uçan, Osman Nuri" seçeneğine göre listele

Listeleniyor 1 - 20 / 114

A bidirectional LSTM-RNN and GRU method to exon prediction using splice-site mapping
(2022) Canatalay, Peren Jerfi; Uçan, Osman Nuri
Deep Learning techniques (DL) significantly improved the accuracy of predictions and classifications of deoxyribonucleic acid (DNA). On the other hand, identifying and predicting splice sites in eukaryotes is difficult due to many erroneous discoveries. To address this issue, we propose a deep learning model for recognizing and anticipating splice sites in eukaryotic DNA sequences based on a bidirectional Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) and Gated recurrent unit (GRU). The non-coding introns of the gene are spliced out, and the coding exons are joined during the splicing of the original mRNA transcript. This bidirectional LSTM-RNN-GRU model incorporates intron features in order of their length constraints, beginning with splice site donor (GT) and ending with splice site acceptor (AG). The performance of the model improves as the number of training epochs grows. The best level of accuracy for this model is 96.1 percent.
A grasshopper optimizer approach for feature selection and optimizing SVM parameters utilizing real biomedical data sets
(Springer London Ltd, 2019) Ibrahim, Hadeel Tariq; Mazher, Wamidh Jalil; Uçan, Osman Nuri; Bayat, Oğuz
Support vector machines (SVM) are one of the important techniques used to solve classifications problems efficiently. Setting support vector machine kernel factors affects the classification performance. Feature selection is a powerful technique to solve dimensionality problems. In this paper, we optimized SVM factors and chose features using a Grasshopper Optimization Algorithm (GOA). GOA is a new heuristic optimization algorithm inspired by grasshoppers searching for food. It approved its ability to solve real-world problems with anonymous search space. We applied the proposed GOA + SVM approach on biomedical data sets for Iraqi cancer patients in 2010-2012 and for University of California Irvine data sets.
A heuristic approach for optimal planning and operation of distribution systems
(Hindawi Ltd, 2018) Alzaidi, Khalid Mohammed Saffer; Bayat, Oğuz; Uçan, Osman Nuri
The efficient planning and operation of power distribution systems are becoming increasingly significant with the integration of renewable energy options into power distribution networks. Keeping voltage magnitudes within permissible ranges is vital; hence, control devices, such as tap changers, voltage regulators, and capacitors, are used in power distribution systems. This study presents an optimization model that is based on three heuristic approaches, namely, particle swarm optimization, imperialist competitive algorithm, and moth flame optimization, for solving the voltage deviation problem. Two different load profiles are used to test the three modified algorithms on IEEE 123- and IEEE 13-bus test systems. The proposed optimization model uses three different cases: Case 1, changing the tap positions of the regulators; Case 2, changing the capacitor sizes; and Case 3, integrating Cases 1 and 2 and changing the locations of the capacitors. The numerical results of the optimization model using the three heuristic algorithms are given for the two specified load profiles.
A hybrid local search-genetic algorithm for simultaneous placement of DG units and shunt capacitors in radial distribution systems
(Ieee-Inst Electrical Electronics Engineers Inc, 2020) Almabsout, Emad Ali; El-Sehiemy, Ragab A.; Uçan, Osman Nuri; Bayat, Oğuz
Controlling active/reactive power in distribution systems has a great impact on its performance. The placement of distributed generators (DGs) and shunt capacitors (SCs) are the most popular mechanisms to improve the distribution system performance. In this line, this paper proposes an enhanced genetic algorithm (EGA) that combines the merits of genetic algorithm and local search to find the optimal placement and capacity of the simultaneous allocation of DGs/SCs in the radial systems. Incorporating local search scheme enhances the search space capability and increases the exploration rate for finding the global solution. The proposed procedure aims at minimizing both total real power losses and the total voltage deviation in order to enhance the distribution system performance. To prove the proposed algorithm ability and scalability, three standard test systems, IEEE 33 bus, 69 bus, and 119-bus test distribution networks, are considered. The simulation results show that the proposed EGA can efficiently search for the optimal solutions of the problem and outperforms the other existing algorithms in the literature. Moreover, an economic based cost analysis is provided for light, shoulder and heavy loading levels. It was proven, the proposed EGA leads to significant improvements in the technical and economic points of view.
A multi-branched hybrid perceptron network for DDoS attack detection using dynamic feature adaptation and multi-instance learning
(Institute of Electrical and Electronics Engineers Inc., 2024) Al-Khayyat, Ali Tariq Kalil; Uçan, Osman Nuri
The increasing sophistication and frequency of Distributed Denial of Service (DDoS) attacks necessitate advanced detection systems. These attacks leave networks vulnerable to disruptions, resource overload, security breaches, and financial losses. Conventional detection systems suffer from high false positive rates, lower flexibility, and an inability to adapt dynamically to trending attack patterns. To address these limitations, our proposed work introduces a novel approach to tackling these challenges by merging a multi-branched hybrid perceptron network with dynamic feature adaptation and multi-instance learning. Our methodology features three key innovations: (1) Multi-Branched Hybrid Perceptron architecture, (2) Dynamic Feature Adaption, and (3) Dynamic Attention-Weighted Feature Fusion to improve feature representation and merging process. The proposed study was validated on three testing datasets: (1) UNSW-NB15, (2) CIC-IDS 2017, and (3) CIC-IDS 2018, and the results were compared with various state-of-the-approaches. The experimental results show that our model significantly outperforms existing methods. On UNSW-NB15, the model achieves an accuracy of 96.02% with a precision of 0.965, a recall of 0.963, and an F1-score of 0.9645. For CIC-IDS 2017, it reaches a near-perfect accuracy of 99.99% with all metrics at 1.00. On CIC-IDS 2018, the model performs with an accuracy of 99.96% and perfect precision, recall, and F1-scores of 1.00. Time complexity analysis shows that while the proposed intrusion detection framework takes 21.6 seconds on CIC-IDS 2017, 30.0 seconds on CSE-CIC-IDS2018, and 15.5 seconds on UNSW-NB15, it remains competitive with high performance. Despite its higher time complexity on UNSW-NB15, MHHPN provides superior detection capabilities, making it practical for real-time use in complicated and extensive networks.
A novel approach for ensuring location privacy using sentiment analysis and analysis for health-care and Its effects on humans health
(Amer Scientific Publishers, 2020) Ali, Bassam S.; Uçan, Osman Nuri; Bayat, Oğuz
Twitter is the free report benefit for given important to as far as possible to 150 typescripts that may include script, pictures, videotapes and hyper-links. Persons sharing news, ideas and info to supports or agonists media. The most petrify theme is the persecution of ladies occurring a rounds the world. Individual's persecution of women of the web-based life to dependably impart utilizing coded words or to setting up their indirections nearness. This paper offering a novel procedure for feelings research on the maltreatment of women connected tweets and to organizing the suppositions with their geo-zones. The information mining calculations, for example, Bolster Vector Machine, Irregular Woodland, Sacking, Choice Trees and Most extreme Entropy are applying for extremity's based characterization of mistreatment of ladies related Tweets. The outcomes are looking at and exhibiting.
A novel energy-conscious threshold-based data Transmission routing protocol for wireless body area network (NEAT)
(Institute of Electrical and Electronics Engineers Inc., 2020) Ibrahim, Abdullahi Abdu; Salawudeen, Ahmed Tijani; Uçan, Osman Nuri; Okorie, Patrick Ubeh
In today's age of wireless communication, Wireless Body Area Network (WBAN) which is an extension of the conventional Wireless Sensor Network (WSN) is attracting immense interest in academia as well as industry. This is due to its importance in providing smart heath care service. One of the major research issues are Quality-of-Service (QoS) provision and energy efficiency improvement. Since sensor nodes are highly resource constrained in terms of battery and it is impractical to recharge and replace them, it is imperative to develop techniques/routing protocols or other solutions in other to augment the battery life. For that reason, NEAT routing algorithm which is an improvement on RE-ATTMPT and CEMob protocols is proposed in this paper. NEAT prioritize data into low-emergency, high-emergency and regular-data. Unlike similar protocols, NEAT ignores the communication of regular-data and transmit high-emergency data via direct communication and low-emergency data is compared with the formerly sensed low-emergency data and if it is different, it is transmitted, otherwise it is not transmitted thus leading to significant energy saving. Simulation results obtained by MATLAB prove that NEAT protocol outperforms RE-ATTMPT and CEMob in terms of network lifetime and throughput. © 2020 IEEE.
A study of scalable solutions for smart contracts performance
(Institute of Electrical and Electronics Engineers Inc., 2024) Alsalim, Mohammed Sabah Hameed; Uçan, Osman Nuri
The emergence of blockchain technology has ushered in a transformative era of decentralized and trustless systems, challenging conventional centralized models reliant on intermediaries. At the core of this revolution lies the concept of smart contracts, self-executing agreements encoded on a blockchain, automating contractual processes without intermediaries. Smart contracts offer numerous advantages, including automation, cost reduction, transparency, and enhanced security. However, their widespread adoption faces challenges, with a paramount one being the limitation of scalability within blockchain networks. This paper studies existing scalable solution of blockchain, compares the existing techniques and discusses the best scalable solution depending on the requirement of the system. It assesses existing research, case studies and unlocking the full potential of smart contracts in the realm of blockchain technology.
A survey on privacy and policy aspects of blockchain technology
(IEEE, 2022) Mahmood, Mohammed Thakir; Uçan, Osman Nuri; Ibrahim, Abdullahi Abdu
From financial transactions to digital voting systems, identity management, and asset monitoring, blockchain technology is increasingly being developed for use in a wide range of applications. The problem of security and privacy in the blockchain ecosystem, which is now a hot topic in the blockchain community, is discussed in this study. The survey’s goal was to investigate this issue by considering several sorts of assaults on the blockchain network in relation to the algorithms offered. Following a preliminary literature assessment, it appears that some attention has been paid to the first use case; however the second use case, to the best of my knowledge, deserves more attention when blockchain is used to investigate it. However, due to the subsequent government mandated secrecy around the implementation of DES, and the distrust of the academic community because of this, a movement was spawned that put a premium on individual privacy and decentralized control. This movement brought together the top minds in encryption and spawned the technology we know of as blockchain today. This survey paper also explores the genesis of encryption, its early adoption, and the government meddling which eventually spawned a movement which gave birth to the ideas behind blockchain. It also closes with a demonstration of blockchain technology used in a novel way to refactor the traditional design paradigms of databases.
A systematic mapping study on touch classification
(Int Journal Computer Science & Network Security-Ijcsns, 2018) Fleh, Saad Q.; Bayat, Oğuz; Al-Azawi, Saad; Uçan, Osman Nuri
One of the basic interpersonal methods to communicate emotions is through touch. Social touch classification is one of the leading research which has great potential for more improvement. Social touch classification can be beneficial in the much scientific application such as robotics, human-robot interaction, etc.. Each person has the ability to interact with the environment and with other people via touch sensors that are speared over human soma. These touch sensors provide us with the important information about objects such as size, shape, position, surface and their movement. Therefore, the touch system plays the main role in human life from early days. The small gesture can express strong emotion, from the comforting experience of being touched by one's spouse, to the discomfort caused by a touch from a stranger. This paper presents and explains a systematic mapping study on social touch gesture recognition. From various digital libraries, 938 papers in total are collected. After applying three filters, 49 papers as primary studies related to the main topic are selected as listed in Appendix (A). The selected papers classified with respect to several facets. The results provide an overview of the existing relevant studies that are reported in the literature, highlight the focused areas and research gaps.
Accuracy enhancement of brain epilepsy detection by using of machine learning algorithms
(Altınbaş Üniversitesi, 2020) Al-Dahhan, Rand Natiq; Uçan, Osman Nuri
Data has gained vital role in science and engineering applications; the proper data analysis has made it possible to boost the economical worthiness of those applications. Machine learning tools are used to classify the big data in order to discover the hidden patterns in them. That may lead to noteworthy advantages that related to future prediction of the data. The resultant information can be used to enhance the practical systems in such way only the profitable thing can be come on then. In other way, it helps to prevent any unpleasant occurrence that may harm the company or the organization. A brain epilepsy disease prediction system is implemented using four different algorithms namely: Naive Bayes algorithm, K-Nearest Neighbours algorithm, Random Forest algorithm and Long Short Term Memory Neural Network. The performance metrics are also initiate in order to evaluate the difference in prediction performance of the four tools. The accuracy of prediction the disease was recorded more likely 33.035, 95, 61.195 and 96.79 for the Naïve Bays, Random Forest, K-Nearest Neighbour and Long Short Term Neural Network.
Achieving optical fiber transmission of over 60 W of electrical power and optical data
(Institute of Electrical and Electronics Engineers Inc., 2023) Al-Hadeethi, Alaa Tareq; Uçan, Osman Nuri; Mohammed, Alaa Hamid
For future mobile communication networks, optically powered remote antenna modules are shown with simultaneous transmission of over 60 watts of electric power and optical data over an optical fibre. As the temperature rises over the set point, the light source undergoes modulation, causing a shift in wavelength. Fibre Bragg Grating Sensors are presented in this study, along with an examination of their design for use in measuring and monitoring powerline temperature change over a certain temperature range. In this piece, we enhance the link design to extract a greater feed light power from the double-clad fibre output, which in turn increases the provided electric power via power-over-fiber connection. Furthermore, we employ a bespoke photovoltaic power converter that directly converts optical power into electric power to increase the electric power for running remote antenna units. The photovoltaic power converter has an efficiency of more than 50% and can absorb feed light of more than 20 W. Up to 43.7 W of electrical power may be sent over the fibre optic cable thanks to the improved design of the power-over-fiber connection and the use of the solar power converter. Power-over-fiber with optical data transfers utilising a single optical fibre has been demonstrated for the first time at this altitude, according to the scientists.
Adaptive video transmission over communication networks
(Altınbaş Üniversitesi, 2019) Göse, Ersin; Uçan, Osman Nuri; Nafea, Ghaidq Nassr; Bayat, Oğuz
This paper is to minimize video bitrate and keeping the high resolution video file manageable. This process is achieved by using new generation of Advanced Video Coders (AVC). Such process can be done at one level encoder or multi levels of encoders by means of transcoding, where reformatting the content to be streamed on channel is called transcoding. The goal is concerned of encoder processes that can be used to achieve transcoding. The codec used for compression of video is H.264, a standard for providing high definition video at substantially low complicity and lower bit rates. The x264 Library is used for encoding H.264 AVC, undergirds some of the most profiles for broadcasting and streaming operations over wired and wireless channels, including different applications. When a technique of bit-rate control is incorporated with the encoder, more reliable and qualified system for low bitrate video streaming over constant bit rate communication channel is achieved, where output rate of the video encoder is controlled by feedback based on the buffer level. Where the most effective parameters such as skip frame, QP, cycle length (Gop), etc. are configured and it used as a rate control tools to test the streaming coded bit rate and the decoded video quality. Testing scenarios use many different videos with QCIF, CIF, and HD formats encoded under main profile. JM19 reference software is used for implementing and testing the standards.
AIoT in healthcare: a systematic mapping study
(Institute of Electrical and Electronics Engineers Inc., 2022) Ibrahim, Hadeel Tariq; Mazher, Wamidh Jalil; Uçan, Osman Nuri
he Internet of Things (IoT) infrastructure and artificial intelligence (AI) technologies are combined to create artificial intelligence of things (AIoT). AIoT in healthcare is a key factor in enabling physicians' offices and hospitals as well as giving patients access to superior scientific facilities. To this end, we conduct a Systematic Mapping Study (SMS) to provide critical information about various applications of AIoT in healthcare. The primary goals of this research are to provide an overview of AIoT research in healthcare and to categorize AIoT research based on annual number of publications, publication venues, and journal distribution. We present some AIoT techniques used in healthcare fields in nine well-known online libraries (IEEE, Springer, Elsevier, MDPI, Taylor and Francis, Hindawi, Wiley online lib., ACM and Google). Initially, we used a number of research questions related to the issue, and when we applied them, 71 studies were produced.
An adapting soft computing model for intrusion detection system
(Wiley, 2021) Alsaadi, Husam Ibrahiem Husain; ALmuttari, Rafah M.; Uçan, Osman Nuri; Bayat, Oğuz
Network security in smart cities has become a key problem in the rapid development of computer networks over the past few years. Intrusion detection systems play a fundamental part in the integrity, confidentiality, and resource accessibility among the multiple network security policies. The classification of the genuineness of packets is main object of the presents research work, the soft computing has applied to classify the genuineness of packets. The complexity of soft computing is greatly reduced if the numbers of features in a dataset are reduced. Managing and analysis of the dimensionality reduction is novelty of the proposed model. The existence of uncertainty and the imprecise nature of the intrusions appear to create suitable fuzzy logic systems for such structures. The neural-fuzzy algorithm is one of the effective methods that incorporate fuzzy logic systems into adaptive and analysis capacities. In this research work, soft computing fuzzy logic system is proposed to enhance network security through intrusion detection. Three network datasets are demonstrated to test and estimate the proposed system. Feature selection has used to remove irrelevant features from entire network data which are obstacle classification processes. The Information Gain method was applied to select importance features for detection intrusion. Adaptive Neuro-Fuzzy Inference System (ANFIS) is further used to process the significant features of the classification network data as normal or attacks packets. Two functions named Jang's Neuro-fuzzy and faster-scaled conjugate gradient (SCG) based on the ANFIS system. Obviously, the experimental results demonstrate the proposed system has attained higher precision in detecting normal or attack. The experimental results have suggested that the proposed system results are better in accuracy and time process for classification compared with the existing models. The Overall Results show that the proposed system can be able to detect various intrusions efficiently and effectively.
An efficient multi-objective memetic genetic algorithm for medical image handling and health safety to support systems in medical internet of things
(Amer Scientific Publishers, 2020) Safi, Hayder H.; Uçan, Osman Nuri; Bayat, Oğuz
Over the most recent couple of decades, the Evolutionary Algorithms (EA) have been considered as a typical dynamic research zone in the field of medicinal picture preparing and wellbeing administrations. This is a direct result of the presence of numerous streamlining issues in this field which can be comprehended utilizing developmental calculations. Besides, numerous certifiable streamlining issues in the restorative picture handling and wellbeing administrations have more than one target work, and for the most part the issue goals are in struggle with one another. The traditional multi-objective transformative calculations perform well when the streamlining issue has less than three goal functions, where the execution of these calculations altogether degrades when the improvement issue has high number of destinations. To deal with this issue, there is a requirement for growing new versatile transformative streamlining mechanisms which can handle the high target capacities in the medicinal picture preparing and wellbeing administrations advancement issues. In this paper, we propose a new developmental calculation dependent on the NSGA-II calculation to productively take care of the many-target enhancement issues. The proposed calculation adds three plans to improve the capacity of NSGA-II calculation when managing with high objective dimensional streamlining issues. Another productive arranging strategy, savvy file and straightforward nearby pursuit are utilized to accelerate the solutions combination procedure to the POF and upgrade the decent variety of the arrangements. The proposed calculation is contrasted and the cutting edge multi-target advancement calculations utilizing five DTLZ test issues. The outcomes demonstrate that our proposed calculation essentially beats alternate calculations when the quantity of target capacities is high. Moreover, we applied our proposed algorithm on medical imaging health problem which is the melanoma recognition problem to enhance the early detection of melanoma disease.
Classification of brain tumors using MRI images based on convolutional neural network and supervised machine learning algorithms
(Institute of Electrical and Electronics Engineers Inc., 2022) Ali, Rawaa; Al-Jumaili, Saif; Duru, Adil Deniz; Uçan, Osman Nuri; Boyacı, Aytuğ; Duru, Dilek Göksel
Brain tumor is abnormal cells that originate from cranial tissue and is considered one of the most destructive diseases, and lead to the cause of death, where the early diagnosis is crucial for accelerating the therapy of brain tumors. Examining the patient's MRI scans is one traditional way of distinguishing brain cancers. The conventional approaches take a long time and are prone to human error, especially when dealing with huge amounts of data and diverse brain tumor classes. Artificial Intelligence (AI) is extremely useful for the strict detection and classification of several diseases in the brain. Convolutional Neural Network (CNN) is one of the modes techniques which act as a tumor classifier due to it shows high effectiveness for diagnosing brain tumors. That's why, in this research, we presented a hybrid method that merged a group of pre-Trained deep learning CNN patterns with a group of supervised classifiers in machine learning called, k-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA). We used an MRI image that consist of images of four brain tumor classes, namely glioma, meningioma, pituitary, and no tumor. We deduced the features extracted from the images by hiring three types of CNN called (GoogleNet, Shuffle-Net, and NasNet-Mobile). Depending upon the experimental consequences, ShuffleN et with SVM achieved the highest results according to the four categories of metrics evaluation that are Accuracy of 98.40%, Precision of 97%, Recall of 96.75%, and Fl-Score of 96.75%. Finally, we compared our results with different state-of-The-Art papers recently published and our proposed method show outperforms compared them.
Classification of resting-state status based on sample entropy and power spectrum of electroencephalography (EEG)
(Hindawi Ltd, 2020) Mohamed, Ahmed M. A.; Uçan, Osman Nuri; Bayat, Oğuz; Duru, Adil Deniz
An electroencephalogram (EEG) is a significant source of diagnosing brain issues. It is also a mediator between the external world and the brain, especially in the case of any mental illness; however, it has been widely used to monitor the dynamics of the brain in healthy subjects. This paper discusses the resting state of the brain with eyes open (EO) and eyes closed (EC) by using sixteen channels by the use of conventional frequency bands and entropy of the EEG signal. The Fast Fourier Transform (FFT) and sample entropy (SE) of each sensor are computed as methods of feature extraction. Six classifiers, including logistic regression (LR), K-Nearest Neighbors (KNN), linear discriminant (LD), decision tree (DT), support vector machine (SVM), and Gaussian Naive Bayes (GNB) are used to discriminate the resting states of the brain based on the extracted features. EEG data were epoched with one-second-length windows, and they were used to compute the features to classify EO and EC conditions. Results showed that the LR and SVM classifiers had the highest average classification accuracy (97%). Accuracies of LD, KNN, and DT were 95%, 93%, and 92%, respectively. GNB gained the least accuracy (86%) when conventional frequency bands were used. On the other hand, when SE was used, the average accuracies of SVM, LD, LR, GNB, KNN, and DT algorithms were 92% 90%, 89%, 89%, 86%, and 86%, respectively.
Classification of single epochs in event related potentials
(Ieee, 2019) Çiçek, Kerime Dilşad; Bayat, Oğuz; Uçan, Osman Nuri; Duru, Adil Deniz
In the concept of this thesis, single trial event related potential measurements were classified. Classification performances of Decision Trees, logistic regression, random forest, Support Vector Machines and XGBoost methods are evaluated. In this context, EEG was collected during the presentation of two different stimulus. The resulting feature set is given as an input to decision trees, logistic regression, random forest, support vector machines, and xgboost classifiers. Due to the limited test data obtained, synthetic Minority oversampling technique(SMOTE) was applied to the data and classification was performed with the updated dataset. As a result of the study, 91% accuracy was obtained for the training dataset in random forest and XGBoost classification methods. For the test set xgboost_tuned has a 62% accuracy and 71% F1 value. To conclude, superior results were found from other classifiers using the xgboost classification method.
Classification of the level of alzheimer’s disease using anatomical magnetic resonance images based on a novel deep learning structure
(CRC Press, 2023) Al-Jumaili, Saif; Al-Azzawi, Athar; Uçan, Osman Nuri; Duru, Adil Deniz
Alzheimer’s is an incurable neurodegenerative disease that generally begins slowly and progresses gradually with time. In the early stage, symptoms of memory loss are mild, while in the late stage it clearly shows the deterioration in cognitive functions. Due to its irreversible nature, early detection reflects positively on reducing restraining the spread and preventing damage to the brain cells thus avoiding reaching the dementia stage. Till now, deep learning is considered to be one of the most significant methodologies used to detect and classify different types of neurological diseases from MRI images. However, in this study, we proposed a novel two-dimensional deep convolutional neural network to classify four stages of Alzheimer’s disease. The dataset consists of four types, namely nondemented, very mild demented, mild demented, and moderate demented subject MR images. First, we have applied a preprocessing technique to resize the image for compliance with our models. Then, we performed Reduce Atmospheric Haze techniques that can decrease the atmospheric haze making all images sharp and clear to feed to the model. We implemented the model 30 times and obtained more than 99.46% for evaluation metrics. The proposed method shows an outstanding performance compared to other papers reported in the literature.

Yazar "Uçan, Osman Nuri" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri