489
Archived Theses
0
DOIs Assigned
0%
DOI Rate
Discipline
Smart location-based mobile shopping Android application
In this thesis, a standard feature of today's smart phones "the location information" is used for determining the user location. SAGO finds nearest electronic super-stores by using the location information. Currently, SAGO shows only the electronic malls however, it will be added other types of shopping stores such as clothing stores, perfumery and so on. Therefore, this mobile shopping Android application's coverage is not limited. The application is easily extendable according to the users' requirements. The Android application is developed in Android 2.2 version, Application Programming Interface (API) level 8. Although the mobile application is developed for Android devices, it will be easily adopted and developed for iOS devices. The system architecture is designed by using Service-Oriented Modeling Framework (SOMF) which is based on the Unified Modeling Language (UML) and also, object-oriented programming language is used in the development process. Designing with UML is simplifying the complexity of the system and helps us to understand the architecture of this thesis project. In the development process, the users' input data which has been brought before searching products in nearest electronic super-stores and also the names of local shopping stores are used in a "Smart Filtering algorithm". Moreover, simple filtering and cleaning technics such as Agglomerative Clustering Algorithm, Greedy Search Algorithm, and a Levenshtein distance are used in Search And Go (SAGO) mobile shopping application. By using this Android mobile application, the users are able to search the products in closer areas. The mobile users do not need to know where the locations of the shopping malls are. Also, the users are able to compare the prices that they are getting from each local electronic super-store and subsequently decide where to buy. The product is not purchased thorough the SAGO mobile shopping application. If the mobile users want to buy products, they will go to the related store directly. According to these, the users save both their money and time by using this Android mobile application.
Indoor Location Estimation by Using Maximum Likelihood Estimation Based Algorithm on Small Cell Networks
This thesis presents a new framework for indoor localization using Third Generation(3G) Universal Mobile Telephone System (UMTS) Femtocell. The fingerprinting technique is applied to collect the Received Signal Strength Indication(RSSI) values through an Android User Equipment(UE) and data is processed in real time using MQTT server. To achieve better Radio Frequency (RF) planning and optimization for the placement of Femto Access Point(FAP), statistical analysis is performed by normalizing and calculating the Mean Squared Error(MSE) of the acquired data. To maximize the success rate in finding the location of the person, Maximum likelihood Estimation(MLE) based algorithm is used for tracking. Simulation was carried out both for 1 million samples and real life test using 100 samples. To make it more accurate and efficient, MLE based algorithm is developed and simulated in MATLAB. Both simulation approaches resulted in good success rates. A case study is also part of this thesis, case study discuss about the real life project and deployment setup for indoor positioning estimation and localization techniques using multiple FAP's.
Evaluation of e-government project in Iraq: the general directorate of traffic (Case study)
The main objective of the e-government project is to provide online services to citizens by using the information technology and particularly the Internet. Thus, the use of information and communication technology (ICT) is the key element of any e-government project. Consequently, great efforts are presented by governments to develop their infrastructure, staffing skills, governmental websites and citizens' awareness to achieve this objective. In Iraq, the government took its first steps to develop the e-government at the end of 2003. These steps represented by putting the implementation strategy of e-government with the help of the United States. Most ministries and institutions had create their own e-government project through their websites. Although it provided few services for citizens, the initiatives were likely good as a young project. Unfortunately, there were many challenges that handicapped the development of the e-government in Iraq involving political reasons, poor infrastructure, digital divide and lack of funding lack. The last reason played an important role in the middle of 2014 and the stakeholder could not maintain the portal of the Iraqi e-government. However, some governmental institutions still create or develop their own e-government projects. This thesis makes an evaluation of the e-government project of the General Directorate of Traffic (GDT) as case study by using a risk assessment technique called Design-Reality Gap (DRG) model to avoid failure in such project we made. This evaluation introduces the gap exist between the design stage and the reality situation in the GDT. Besides, it takes actions to reduce the gap to increase the success rate. In addition, a proposed assessment technique used to complete the evaluation process. This thesis also presents the e-government of Iraq in general as well as the e-government of two other countries. During the evaluation process, the thesis emphasizes on the technical aspects of the e-government project.
Estimation of commercial building energy consumption with machine learning
The estimation of energy consumption in commercial buildings holds immense significance in the pursuit of sustainable energy management and the efficient allocation of resources. As energy demands continue to rise, optimizing energy consumption becomes crucial for reducing carbon footprints and enhancing cost-effectiveness. In response to these challenges, this research endeavors to leverage the power of machine learning (ML) techniques to accurately predict energy consumption in commercial buildings. By employing ML algorithms, this study seeks to improve energy efficiency, lower operational costs, and facilitate informed decision-making in building energy management. To achieve these objectives, the research employs eight ML algorithms, namely Support Vector Machines (SVM), Random Forest (RF), Extra Trees, Linear Regression, Lasso, Gradient Boosting Regressor (GBR), Multilayer Perceptron (MLP), and a novel stacking model. Each of these algorithms is well-known for its predictive capabilities and ability to handle various types of data. The research methodology encompasses the development and assessment of predictive models using a robust and extensive dataset, carefully collected from commercial buildings. Before applying the ML algorithms, the dataset undergoes rigorous preprocessing to ensure data quality and enhance model performance. Normalization, data cleaning, and feature selection techniques, such as ANOVA and Relief F, are employed to eliminate noise and irrelevant information, making the dataset suitable for training and evaluation. The research splits the data into training and testing sets in a balanced ratio of 70:30, ensuring that the models are trained on a substantial portion of the data while being evaluated on unseen samples. ii This balanced data splitting allows for unbiased evaluation and comparison of the performance of the ML algorithms. Remarkably, the results reveal that the stacking model, a novel approach that synergizes multiple models' strengths, outperforms all other ML algorithms and state-of-the-art approaches. The stacking model showcases an impressive Mean Absolute Error (MAE) of 0.01286, Mean Squared Error (MSE) of 0.00054, and Root Mean Squared Error (RMSE) of 0.02328. These exceptional performance metrics demonstrate the stacking model's potential as a highly effective tool for accurately estimating energy consumption in commercial buildings. The research findings are further validated by comparing the results with related works in the field. The stacking model consistently outperforms existing approaches, reaffirming its superiority in accurately predicting energy consumption. This model's success can be attributed to its ability to combine diverse models, leveraging their complementary strengths to capture complex relationships within the energy consumption data. The adoption of the stacking model as the best-performing ML algorithm in energy consumption estimation carries significant implications for sustainable energy practices in the built environment. Accurate energy consumption prediction empowers building managers to make informed decisions on energy allocation, implement energy-saving measures, and optimize energy usage. By enabling data-driven energy management, the stacking model supports the advancement of sustainable energy practices, contributing to the reduction of carbon emissions and the promotion of environmentally friendly building operations. In conclusion, this research showcases the immense potential of machine learning techniques in accurately estimating energy consumption in commercial buildings. The stacking model's exceptional performance, when compared with other ML algorithms and state-of-the-art approaches, highlights its role as a transformative tool in achieving energy efficiency and sustainable energy management. As the world continues to prioritize sustainability, the implementation of the stacking model can significantly contribute to building a greener and more energy-efficient future.
User behavior analysis on e-commerce using NLP techniques
This study presents an in-depth investigation into the potential for sentiment analysis (SA) and machine learning (ML) in facilitating the sales prediction and customer retention processes for both small and large-scale businesses. Online platforms including blogs, social networks, and review portals have transformed the marketing landscape, enabling consumers to voice their opinions on a vast array of topics, from product reviews to popular culture. These digital platforms not only foster customer engagement, but also provide businesses with an invaluable data source for predictive analysis, essential in strategic sales forecasting and customer relationship management. In this study, we compiled a comprehensive dataset of product review tweets, serving as a rich representation of consumer sentiment. To ensure the integrity and relevance of the data, we engaged in rigorous preprocessing methodologies, mitigating potential noise and inconsistencies. Following the cleaning phase, we utilized the Valence Aware Dictionary and sEntiment Reasoner (VADER), a lexicon and rule-based sentiment analysis tool, in conjunction with several machine learning algorithms. The objective was to ascertain the most effective means of classifying the sentiment polarity of product reviews, subsequently aiding in sales prediction. Our findings reveal that while VADER offers notable benefits in sentiment analysis, ML techniques present superior accuracy in classifying the polarity of product reviews. More specifically, logistic regression (LR) was found to be the top-performing algorithm in this context. Across a multitude of evaluation metrics, including accuracy, precision, recall, F1-score, Matthews correlation coefficient (MCC), and area under the curve (AUC), LR consistently outperformed its ii counterparts, thus solidifying its position as the most apt choice for sentiment classification in product reviews. Notably, other algorithms such as XGBoost and Stochastic Gradient Descent (SGD) also demonstrated competitive performance. They can serve as plausible alternatives in situations where model interpretability is not the prime concern and a higher degree of model complexity is permissible. These findings contribute to an emerging body of knowledge, illuminating the potential of SA and ML in providing businesses with robust tools for understanding customer sentiment, predicting sales, and consequently enhancing customer retention strategies. The implications of this study extend beyond academia, promising substantial real-world benefits for various stakeholders in the business sphere.
Predicting the value of football player with the impact of covid-19 on the market value of the player
The unprecedented onset of the Covid-19 pandemic has pervaded countless industries worldwide, with football being a major one to bear its brunt. This global health crisis ushered in a new era where normalcy was challenged and socio-cultural constructs were altered, leading to profound implications on football, both as a sport and an industry. Football, a nexus of economic and cultural intersections, witnessed several changes, not least among which was the effect on player valuations in the market. The pandemic brought life to a standstill, ensnaring societies in its clutches and forcing them into isolation. This sudden cessation had a ripple effect on football players, amplifying feelings of anxiety, tension, and uncertainty, thus altering their psychosocial dynamics. Consequentially, this shift in psychological state, combined with the disruption of regular football activities, contributed to fluctuations in the market prices of these athletes. Traditional approaches and methodologies, once relied upon for gauging player prices, particularly in the five major European football leagues (English, Spanish, Italian, German, and French), began showing signs of obsolescence in the face of these pandemic-induced challenges. So, this study presents the performance assessment of five machine learning algorithms—Linear Regression, Lasso Regression, Random Forest, Gradient Boosting, and K-Nearest Neighbors—on a regression problem. The evaluation metrics utilized are Mean Absolute Error (MAE), Root Mean Absolute Error (RMAE), and R-squared (R2). Linear Regression demonstrates a high R2 of 0.97 but falls short in MAE and RMAE. Lasso Regression surpasses with an R2 of 0.99 but exhibits the highest prediction errors, hinting at possible overfitting. Random Forest provides a balanced outcome with a R2 of 0.91 and moderate errors. Gradient Boosting stands out in terms of prediction accuracy, having the smallest MAE and RMAE, despite an R2 of 0.90. Meanwhile, K-Nearest Neighbors parallels Linear Regression's performance. The study suggests that Gradient Boosting might be the most accurate for predictions based on MAE and RMAE. However, the ideal model selection is contingent upon the specific objectives of the regression problem, which could prioritize explaining variance or minimizing prediction errors.
Gromacs tabanlı MD simülasyon hazırlığına yönelik arayüz geliştirilmesi
Moleküler dinamik (MD) simülasyon, biyolojik ve kimyasal süreçlerin atomik düzeyde anlaşılmasında kritik bir yöntemdir. Ancak bu simülasyon, özellikle GROMACS gibi yazılımlar kullanılarak yapıldığında hem yazılımsal bilgi ve beceri hem de yüksek hesaplama gücü içeren donanım gerektirir. Bu durum, birçok araştırmacı için erişim engeli oluşturmakta ve bilimsel süreçleri yavaşlatmaktadır. Bu çalışmada, GROMACS tabanlı simülasyon süreçlerini kullanıcı dostu bir web ve bulut tabanlı platform ile otomatize ederek, simülasyon erişimini kolaylaştırmayı ve kullanıcı dostu bir arayüzün üretilmesi sağlanmıştır. Sistem ile kullanıcıların protein ve ligand dosyalarını yükleyerek simülasyonu başlatmaları, sonuçları görselleştirmeleri ve elde edilen sonuçların paylaşımı mümkün kılınmıştır. Böylece hem yazılımsal uzmanlığı sınırlı kullanıcıların bu alana girmeleri teşvik edilmiş hem de deneyimli araştırmacıların verimliliği artırılmıştır. Proje, web arayüzü, lokal GROMACS sunucuları ve bulut tabanlı bir altyapıyı bir araya getiren bir yöntemle yürütülmüştür. Yazılım web arayüzüne yüklenen dosyalar, GROMACS sunucularına iletilerek internet üzerinden gerekli ek dosyaların indirilmesi ve standart GROMACS iş akışı yürütülmesi ile yürütülmüştür. Simülasyonun tamamlanmasının ardından, sistem otomatik olarak RMSD, enerji grafikleri ve atomik mesafe değişimleri gibi grafiksel çıktılar üretilmiştir. Bu çıktılar, bir bulut sunucusuna yüklenerek kullanıcıların sonuçlara kolayca erişimi hem web üzerinden hem de e posta yoluyla sağlanmıştır.
Low cost photovoltaic monitoring system based on lorawan network
Renewable energy, especially solar energy, is increasingly popular in both home and industrial sectors due to globalization. This thesis investigates the design and implementation of a low-cost photovoltaic (PV) monitoring system using the LoRaWAN network. Because it offers a virtual environment for modelling, evaluating, and experimenting with complex systems, the simulation of monitoring a PV system is essential for comprehending system behaviour. The system seeks to provide a low-cost solution for monitoring energy and environmental parameters in PV systems by using the features of LoRaWAN technology for long-range, low-power communication. This thesis offers a thorough analysis of the application of Internet of Things technology for the real-time monitoring of PV system performance. In order to improve the overall efficiency of PV systems, the goal is to offer a dependable and scalable system that facilitates ongoing monitoring, data collecting, and analysis. The system addresses the issues of cost-effectiveness, scalability, and remote monitoring in PV systems by including sensors to capture vital data, such as solar panel output, battery condition, and ambient variables. A LoRaWAN network, which offers long-range and low-power communication and supports data transfer, is perfect for off-grid and remote photovoltaic systems. By adding LoRaWAN technology, systems become more efficient and scalable, opening the door for widespread use of sustainable energy solutions. PV systems are now affordable enough for a broad spectrum of customers, including those who live in remote or underdeveloped areas. The system offers real-time data on PV system performance, facilitating preventive maintenance and well-informed decision-making, according to the findings of the field testing. The data from the DC voltage sensor and PZEM-004T V3 module provide crucial insights into the dynamic behavior of the electrical system under observation. The system's need for electrical power is implied by the continuous increase in active power consumption (Pac), which went from 933 to 2107 Watts. In the MATLAB/Simulink system, the measurement results are concurrently entered into the PV model enabling theoretical simulation and graphically shown as a dashboard. On the same software platform, the assessment and defect detection functionalities for PV modules within real-world operating circumstances are carried out and presented. The capabilities of visual monitoring, assessment, and defect detection of the suggested system have been shown with adequate confidence and accuracy. The suggested system has several benefits over the well-developed ones, including the ability to combine all visualisation monitoring, assessment, and fault detection inside the MATLAB/Simulink environment and to decrease both wire and hardware configuration.
Streamlining life certificate authentication: Biometric integration with fingerprint and vein recognition
With the rapid advancement of technology worldwide, there is an increasing demand for innovative solutions to address challenges in managing and delivering public services, particularly for pensioners. One key challenge faced by institutions responsible for disbursing financial entitlements to pensioners is the "proof of life" verification process, which is often time-consuming and prone to errors. This thesis proposes a digital solution that enhances security and accuracy by combining two biometric authentication methods—fingerprint and vein recognition. By integrating these complementary biometric techniques, the system significantly improves the accuracy of the "proof of life" process. Experimental results demonstrated that the dual-authentication system achieved higher accuracy compared to using a single method, providing over 87.07% accuracy in verifying pensioners' identity, with verification times under 4 seconds. This efficient and secure approach ensures that eligible pensioners can continue receiving their pensions without delays, benefiting both pensioners and service providers.
Optimizing photovoltaic system diagnostics: integrating machine learning and DBFLA for advanced fault detection and classification
Because of the exponential development in the number of photovoltaic (PV) power plant installations, standard inspection procedures have become ineffective. As a consequence of this, there is a requirement for more sophisticated approaches to the identification and categorisation of faults. In order to do this, the DBFLA method, which is a unique hybrid metaheuristic technique, will be presented in this paper. Based on the Dung Beetle Optimisation Algorithm and Fick's Law of Diffusion Algorithm, this technique is aimed to answer the issues that have been highlighted. It is a hybrid of the two algorithms. The DBFLA improves the efficiency of machine learning models such as ANN, SVM, and ensemble approaches by modifying their parameters in order to improve the precision of fault detection. This is accomplished through the process of adjusting the parameters. The identification of problems such as open circuits, short circuits, and module incompatibilities may be accomplished in a rapid and precise way. In accordance with the results of the study, DBFLA is able to construct a stacking classifier by making use of actual PV datasets. This classification strategy achieves an individual meta-learner accuracy of around 98.75%, which is a substantial improvement above the performance of standard machine learning approaches. The capacity of this technique to handle a bigger number of operating modes and a broader diversity of issue scenarios has resulted in the development of enhanced fault detection systems. This evolved as a subsequent consequence of the method's ability to accommodate these capabilities. The improvement in classification accuracy is the most important contribution that DBFLA brings to the table, in contrast to the optimisation procedures that have always been used before. The capability of the approach to properly strike a balance between exploration and exploitation is directly responsible for this advancement on the part of the method. This hybrid approach is provided in order to demonstrate how actual and simulated information may be merged to bring about PV defect detection algorithms that are both more accurate and more efficient. The goal of future research is to improve the operational efficiency and reliability of photovoltaic (PV) systems by incorporating these complicated models into real-time monitoring systems. This will be accomplished through the adoption of advanced modelling techniques.
The role of machine learning in enhancing realism in unreal engine games
The way machine learning (ML) is applied in game production process, especially through development IDE such as the Unreal Engine 5, paves innovative methods for developing significant immersive and authentic digital experiences. This thesis inspects the relational intersection of machine learning and game development process with examination of spatial how machine learning methods lends to gameplay mechanics, player interactions, and simulates physics-based movements. Additionally this thesis also contains a case study about movement mechanics of wind turbines in a simulated environment, using Unreal Engine's blueprint scripting way of programming but also hardcode in C++ to make blueprints run into Unreal Engine's blueprint scripting. Used coded simulation using real time Machine Learning algorithm on C++ to calculate the turbine's body and blade movement and simulated it with real world factors such as wind source, distance and speed. This calculation impacts the amount of power that is produced by the wind turbine. This tells that applying these methods and technologies with all of ML algorithms, Unreal Engine 5's in engine physics simulations and time may enhance visual environments, fidelity and deep game environment mechanics. This work demonstrates that ML based simulations can be done in real time in a way that reflects player behaviors and environmental influences in the game thereby improving the gameplay experience. The game utilizes AI-driven simulations that respond to player behavior, as well as dynamic interactions with environmental changes, contributing to greater immersion. Finally, this research explores finds and includes the challenges for importing real time Machine Learning algorithms into games, uncovering the potential resource usage and performance optimizations, leads and provides insights into to potential future of advanced ML and AI powered game designs, mechanics and coding. This thesis and project provides an improvement in knowledge on using the ML and AI algorithms into the gaming part of life and innovative thinking of adapting these technologies for realism techniques to extend the limit of your in-game interactions and not Static experiences in virtual fronts.
A new IoT security framework using hybrid deep learning techniques
Connecting systems, apps, data management, operations, the Internet of Things builds a network, continuously supports organizations while also opening up new avenues for cyberattacks. IoT security is currently seriously threatened by illicit downloading and virus attacks, which have the potential to compromise private data and harm a company's reputation and finances. Here we describe a hybrid deep learning optimization strategy for detecting and averting assaults in Internet of Things environments. We build a cybersecurity warning system index by first identifying and quantifying pertinent aspects, and then we assess the situation. We employ bio-inspired approaches to increase the efficiency of an Intrusion Detection System (IDS) by lowering the dimensionality of the data and eliminating noisy inputs. One such method that improves IDS effectiveness is the Grey Wolf Optimization (GWO) algorithm, which can identify both typical and anomalous network congestion. By using different pre-processing techniques, we have enhanced the intelligent initialization step and made sure that informative features are there right away. To minimize underlying data characteristics in a large data environment, To find and confirm index components, integrate Whale and Grey Wolf optimization with a deep learning strategy in simulation to prevent attacks. TensorFlow is a deep neural network that uses system software plagiarism detection to classify software that has been copied. Our suggested strategy for assessing cybersecurity threats in IoT offers better classification results than current approaches, as evidenced by experimental data. Therefore, we use Whale and Grey Wolf Optimization (WGWO) in combination with a deep convolutional network for efficient attack avoidance in IoT.
An eclectic design and adas-oriented realtime simulation of AEB with adaptive brake algorithm under Euro NCAP protocol
This thesis presents the development and evaluation of an adaptive braking algorithm integrated into an Autonomous Emergency Braking system, designed in accordance with Euro NCAP scenario protocols. The primary objective is to improve collision avoidance performance and driving comfort compared to classical fixed deceleration models. A total of seven critical driving scenarios including car to car, pedestrian, and bicyclist interactions were modeled using MATLAB/Simulink. The simulations were conducted under three different road surface conditions: dry, wet, and icy. While the classical braking model applies pre-defined deceleration values once the AEB system is triggered, the proposed adaptive model continuously calculates and applies optimal braking force in real time, based on factors such as time to collision, ego vehicle speed, relative distance, target speed, and road friction coefficient. The results show that the adaptive braking system achieves full collision avoidance in all tested scenarios and surfaces. It demonstrates a clear advantage in avoiding unnecessary braking, maintaining passenger comfort, and minimizing stress on the vehicle's mechanical components. The adaptive model also provides better responsiveness in complex or low-traction environments compared to the fixed model. This research highlights the importance of real-time, environment-aware braking strategies for next-generation Advanced Driver Assistance Systems, and suggests that data-driven AI-based decision models could further enhance the flexibility and reliability of emergency braking systems.
Advanced 3D face anti-spoofing system using hybrid deep neural network and optimization techniques
ace recognition technology is used everywhere, including mobile security, surveillance, and online payments, to authenticate a person's identity. Spoof faces, like printed photos, videos, 3D masks, and deep fake-generated faces, can mislead such systems. The goal of this research is to improve Face Anti-Spoofing (FAS) techniques to improve facial recognition security, accuracy, and efficiency. The study suggests a 3D Face Anti-Spoofing Model incorporating Dense Squeeze and Excitation Networks and a Neighbourhood-Aware Kernel Adaptation (NAKA) process for identification of fine details and textures on a person's face. A Lightweight Multi-Modal Deep Fusion Network is suggested for fusing different face data modalities such as RGB images, depth maps, and texture information. It is useful in face spoofing detection with higher accuracy even against advanced attacks. Deep Reinforcement Learning (DRL) is also used by the system to learn automatically and keep itself updated at all times, thus able to identify new ways of spoofing. The models are tested using standard datasets like CASIA-SURF, CelebA-Spoof, and GREAT-FASD-S based on a number of performance metrics such as accuracy, precision, recall, and error rates. Experiments demonstrate that models as proposed compare favorably against existing methods and are computationally lightweight for use in real-world applications. The study also confronts fundamental problems such as generalizable good performance over datasets, adversarial attacks resistance, and security vs. usability. Grounded on multi-modal fusion, vi attention, and reinforcement learning, the study gives contributions towards robust, light-weight, and explainable face anti-spoofing approaches. These can be applied in high-security applications like banking, border control, and mobile authentication. The outcome will help to introduce biometric security and protect against future face spoofing attacks of novel forms.
Kantitatif hisse senedi analiz uygulaması
Bu tez çalışmasında, Kamuyu Aydınlatma Platformu (KAP) ve Türkiye Bankalar Birliği (TBB) gibi resmî kaynaklardan elde edilen çoklu finansal verilerin, ASP.NET Core tabanlı katmanlı bir mimari üzerinden Playwright ve Selenium ile zamanlanmış veri çekimi, otomatik ön işleme, SQL veri tabanı kayıt altyapısı ve Python destekli görselleştirme bileşenleri aracılığıyla entegre edildiği uçtan uca otomatik bir kantitatif analiz sistemi geliştirilmiştir. Sistem hem reel sektör hem de finans sektöründeki kurumlara yönelik olmak üzere Kantitatif, Banka, Sigorta ve Faktöring başlıklı dört farklı PDF raporu üretmekte; ayrıca günlük piyasa özetlerini oluşturmakta ve Telegram bot entegrasyonu aracılığıyla gerçek zamanlı bildirimler sağlamaktadır. Elde edilen deneysel bulgular, çok kaynaklı veri entegrasyonunun analitik kapsamı anlamlı ölçüde genişlettiğini; Excel ve PDF modüllerinin ise raporların detay seviyesini önemli ölçüde artırdığını göstermektedir. Performans ölçümleri, sistemin raporlama süresini %65'e kadar azalttığını ve sektör bazlı rating karşılaştırmaları sayesinde portföy kararlarında duyarlılığı ve doğruluğu artırdığını ortaya koymuştur. Tartışma bölümünde veri güncellemelerinde senkronizasyon sorunları, zaman damgası uyumsuzlukları ve versiyon kontrolü gibi çoklu kaynak yönetimi zorlukları analiz edilmiştir. Sistem mimarisi, API tabanlı servisler, gömülebilir bileşenler ve çoklu kiracı yapısı sayesinde, ticarileştirilmeye uygun tam ölçekli bir SaaS (Software as a Service) ürünü niteliği taşımaktadır. Sonuç olarak geliştirilen sistem; veri doğruluğu, işlem hızı ve kullanıcı etkileşimini tek bir platform altında birleştirerek, Türkiye finansal piyasaları için ölçeklenebilir, sürdürülebilir ve genişletilebilir bir karar destek aracı sunmaktadır.
Derin öğrenme algoritmalarına dayalı ağ performansını iyileştirmek için ıot ağlarının eğitimli izlenmesi
The introduction of the IoT has forced the integration of billions of devices in different sectors, thereby creating huge data and change. Nevertheless, IoTs have some few challenges when it comes to internetworking and securing the networks. Networking problems like latency, packet loss, congestion and probable security holes make it imperative that networking headers are monitored and monitored are checked for anomalies. This thesis proposes a deep learning-based approach to real-time IoT network monitoring and anomaly detection, focusing on three models: FFNN, CNN, and MLP are the popular categories of Deep Learning Algorithms. The models were created and built with MATLAB to review IoT network data to identify discrepancies, distinguish malfunctioning nodes, and diagnose future problems. In an effort to enhance the outcome of the models, optimization methods of Adam and Stochastic Gradient Descent with Momentum (SGDM) was used. The models were tested on synthetic IoT data and the results highlighted by using the quality control indicators such as accuracy, precision, recall, and F1-score. However, the results show the proposed methods improve the existing results where MLP and CNN have higher accuracy and anomaly detection rates than FFNN, MLP-93 (92.3%), CNN-94 (94%), DT (78.5%), and SVM (85.7%). This considerable enhancement is due to efficiency of CNN for extracting spatial relationship and MLP to learn non-linear relationship in IoT network data. Comparing with other methodologies implemented in the current research, deep learning models provide a higher level of accuracy in the identification of sophisticated abnormalities, which is beneficial for real-time IoT monitoring. In this case, the CNN model showed remarkable improvement on its capability in the identification of network patterns and the prediction of issues as compared to prior models. These outcomes give a strong signal that deep learning models, including CNN and MLP, are more beneficial for real-time IoT network performance monitoring and anomaly identification than conventional models of machine learning. The work from this research can be generalized as follows to be used in improving the dependability and securability of the IoT networks while giving a much better solution than the current method.
Deep learning-driven classification ofdepth image-based rendering anddynamic face warping in deepfakealtered videos for fabricated news onsocial media
Deepfakes present a persistent challenge in a world driven by transparency and truth. This study introduce a new deepfake detection method based on Xception architecture. Deepfake is having some threats to the authenticity of digital information especially on social media platforms through manipulated content, which transpires fast within these networks and powerfully influences the public's opinion, reputational harm, and fuel of misinformation. This research introduced an Xception-based approach to deepfake detection. The architecture utilizes an efficient feature extraction CNN model that benefits from the improvements such as depthwise separable convolutions and is optimized to exploit subtle artifacts in typical deepfake media, such as inconsistencies in facial landmarks, lighting irregularities, and unnatural textures. It was trained and tested on an all-inclusive dataset of more than 500,000 frames comprising authentic and manipulated video content. The data preprocessing steps included face extraction and alignment and diversity augmentation of the data to make generalization better. The model was evaluated using various metrics and scored quite impressively at 99.69% accuracy, precision at 99.58%, recall at 99.80%, and with an F1 score at 99.69%. These results demonstrate that the model is correctly balanced to avoid false positives at a reasonable tradeoff for identifying deepfake content well enough to be deployable in real-world scenarios where content verification will prove critical. The Xception-based model also shown strong computational efficiency with the average inference time being 0.08 seconds per frame. With an AUC-ROC score of 0.9999, this efficiency combined proves that the model can significantly distinguish the real from the manipulated frames with near-perfect accuracy. More so, its robustness was tested under various conditions, including different resolutions and compressions. It achieved an accuracy of 99.65% on low-resolution frames and 98.1% on high-resolution frames, with an accuracy of greater than 90% even for highly compressed videos that ascertained its adaptability to the diverse media qualities characteristic of content usually found on social media. In summary, a deepfake detection framework based on Xception offers a powerful, accurate, and efficient solution to deepfake media, supporting the integrity and authenticity of digital content on social media. As possible future directions of this work, it could be considered to further explore the adversarial training for more robustness against emerging deep fake generation techniques and expanding the dataset with occluded and diverse samples. This research is therefore a valuable contribution to the area of deep fake detection, offering scalability of the tool to reduce misinformation and protect public trust in digital media.
Ai-based optimization of blockchain protocols for cybersecurity in iot systems in medium-sized organizations
The Internet of Things (IoT) has enabled small and medium enterprises (SMEs) to automate and make their operations more efficient. Unfortunately, the need to keep these facilities online exposes them to cybersecurity threats that can wreak havoc on their reliability and integrity. The object of this study is to fill this gap by assessing AI-driven optimizations and their effects on real-time threat detection and response. Data regarding enterprise IoT utilization and customers' behavior were collected and analyzed using statistical methods such as percentage distributions, normality tests, linear regression, analysis of variance (ANOVA), etc. The improved outcomes in cybersecurity of AI-enhanced blockchain protocols were validated using the Amazon IoT dataset, yielding a relative coefficient R² of 0.947. AI also played a sig. positive role in real-time threat detection and mitigation, backed by p-values (<0.05) and statistical sig. (>2). These results place emphasis on the continuous training of AI models and regular updates of blockchain protocols as mandatory for the sustainability of the system, resilience, and ability to withstand multi-pronged and persistent threats.
Model predictive control for passing assistance in autonomous vehicles
Autonomous or self-driving vehicles are designed to operate without the need for direct driver control over steering, acceleration, and braking. These vehicles are intended to function in self-driving mode without requiring the driver to continuously monitor the road. The aim of this thesis is to present a novel and robust methodology contributing to the improvement of the performance of autonomous vehicles in both simple and complex manoeuvres. In this context, the dual-controller approach is utilized for merging the benefits of a Model Predictive Controller with a Stanley controller into a hybrid system, namely the Model Predictive and Stanley-based Controller (MPS). Each of them is suffering from some shortcomings: MPC can efficiently perform the path prediction but may lag in responding properly under dynamic conditions, while Stanley controller-though widely used for lateral control-exhibits high lateral errors on tight curves and complex road structure. These are combined carefully in the MPS method so that their respective disadvantages counterbalance each other: the predictive capabilities of MPC are combined with the robustness of the Stanley controller to yield superior path-following and vehicle control capabilities. Extensive analysis of the MPS system will thereafter be implemented to evaluate its performance under a variety of road conditions, namely straight sections, tight corners, and road environments filled with obstacles. The controller that has been developed demonstrates a level of flexibility and attainable adaptability that is appropriate for a wide range of scenarios while retaining lane stability and trajectory accuracy throughout a variety of speeds and road categories. Accordingly, the several measures of performance for tracking accuracy, minimization of errors, and computational efficiency indicate that the MPS system surpasses the single controller approach system, especially in situations where other systems may face difficulties. The paper also performs a comparison analysis with traditional control systems, pointing out that the MPS controller gives better performance in minimizing errors and providing smoother and more reliable vehicle movement under simple and complex conditions. Conclusively, the research positions the MPS controller as a significant advancement in the field of autonomous vehicle control by proposing an innovative combination of advantages of the model predictive and Stanley controllers. The results obtained so far have shown that MPS can help achieve safer and more precise autonomous driving, thus making it a potential candidate for real-world applications where reliability and adaptability are considered of prime importance.
Backscatter-assisted non-orthogonal multiple access for ultra-massive machine-type communications in 6G networks
6G cellular networks will require to provide significantly higher system capacity and user data rates. This potential growth along with today's shortage of spectrum increases the need for new frequency spectrum. The new millimeter wave spectrum is emerging as a suitable candidate with a large amount of available bandwidth (around 60 GHz). The new spectrum places a new requirement for single element antenna and array design. This work addresses the issues of mm-wave antenna design and the problems that designer may face during the designing process. The proposed antenna has a resonating frequency of 26 GHz and 2 GHz bandwidth. Beam squint problem is also analysis in this work. The results showed that the gain of the mm-wave antenna array becomes a function of frequency which significantly reduces the performance of mm-wave communication system. Millimeter wave (mmWave) wireless technology has become a part of human life for high-speed and secure data transmission. A square microstrip patch antenna with a resonance frequency of 26GHz was proposed in this work for mmWave wireless communication. One square radiating element makes up the antenna. CST Microwave Studio, an electromagnetic simulation program, was used to construct and study the suggested antenna on a Rogers RO 3003 lossy substrate with a relative permittivity of 3. This work's outcome demonstrates a minimal return loss of -19.34 dB, a gain of 6.97 dBi, and a bandwidth of 2GHz at a resonant frequency of 26 GHz. The element is transformed into an 8-element uniform linear array; the suggested array boosts the gain to 16dBi while maintaining a high radiation efficiency. A high-gain wide-band planar antenna with a reconfigurable intelligent surface (RIS) is presented in this study for use in contemporary wireless communication applications. The antenna is composed of two primary components: a basic antenna component with two light-dependent resistor switches and cross-line slots, and a second component that uses the RIS layer for beam steering. The RIS is made up of two-sided, five-by-five-unit cells that form a square. The antenna substrate is a 1.6 mm thick dielectric layer of FR4 epoxy glass. In order to attain the appropriate electromagnetic properties at the frequency band of interest, the RIS inclusions are developed and numerically tested. The manufactured prototype achieves an antenna gain ranging from 10.5dBi to 16.8dBi and exhibits a wide band covering frequencies from 0.9GHz to 3.5GHz with S11 below -10dB. Effective aperture utilization is demonstrated by experimental measurements in all shapes, and beam steering from +22° to -22° is achieved without lowering side-lobe levels. To assess channel performance in terms of bit error rate (BER) and channel capacity (CC), the suggested antenna's performance is compared to actual measurements. In contrast to traditional RIS-assisted antennae that depend on PIN or varactor switches, the presented work conducted LDR-controlled design offers compact beam steering with little insertion loss.
Enhanced energy demand management in electric distribution networks using lstm-xgboost model
Accurately forecasting electricity consumption is essential for improving the reliability, operational efficiency, and resilience of energy systems, particularly given the rising integration of renewables and the advancing complexity of smart grids. This research introduces an innovative multilayered model for short-term electric load forecasting that combines Long Short-Term Memory (LSTM) networks with Extreme Gradient Boosting (XGBoost) to enhance accuracy and dependability. The approach consists of three main steps: data cleaning and feature extraction, development of separate LSTM and XGBoost models, and combining them into a unified hybrid architecture. The Elia Grid dataset from Belgium was used in this study, containing high-resolution load data for 2022 captured at 15-minute intervals. The hybrid model leveraged the LSTM's strength in learning sequential dependencies, while XGBoost contributed by capturing non-linear residual patterns and extracting feature importance. The proposed model underwent extensive testing to evaluate its performance against independent LSTM and XGBoost models. The hybrid model achieved its best results with a Root Mean Square Error (RMSE) of 106.54 MW and Mean Absolute Percentage Error (MAPE) of 1.18% and Coefficient of Determination (R²) of 0.994. The sensitivity analysis showed that increasing the look-back window size improved model performance but the attention mechanisms did not enhance accuracy so they were removed from the final design. In addition to outperforming traditional models, the proposed framework demonstrated strong generalizability, scalability, and interpretability, making it suitable for real-time energy management systems. The model was benchmarked against recent classical and deep learning (DL) models and showed competitive or superior results across multiple datasets. This work contributes to the design of AI-based applications in smart grid management and enriches the literature on hybrid DL techniques for time series forecasting. Future research will explore expanding the model to multi-regional datasets, incorporating weather and socioeconomic variables, and deploying the system in real-world grid control environments.
Yakın kızılötesi ışık spektroskopisi tekniği ile kan glikoz yoğunluğu değişiminin gözlenmesi
Diyabet hastalığı kandaki glikoz konsantrasyon değerinin vücut tarafından dengelenememesinden dolayı ortaya çıkan metabolik bir rahatsızlıktır. Bu sebeple diyabet hastaları hastalığın komplikasyonları ile karşılaşmamak için kan glikoz konsantrasyon değerini düzenli olarak takip etmeleri gerekmektedir. Bu takip işlemi 1970'lere kadar evlerde yapılamazken gelişen teknoloji ile birlikte evlere girmesi mümkün olmuştur. Kullanılan glikoz konsantrasyon takip sistemleri sırasıyla invazif, minimal invazif ve non-invazif tekniklerle ölçümler yapılmaya çalışılmıştır. Birinci nesil ve ikinci nesil ölçüm sistemlerinde invazif yöntemler kullanılmıştır. Fakat invazif yöntemle yapılan ölçümler hastalar üzerinde rahatsızlık vermesinin dışında enfeksiyon kapma riski taşımakta ve ölçüm bölgesinin sinirlerinin ölmesine sebep olduğundan yapılan çalışmalar non-invazif yöntem üzerinde yoğunlaşmaktadır. Şimdiye kadar non-invazif ölçümler için bir çok teknik kullanılmış ve sayısız çalışma yapılmıştır. Bu yöntemlerin bazılarıyla klinik olarak kısmen hassas ölçümler elde etmeyi başarılmıştır. Fakat yapılan çalışmalarda olumlu sonuçlara en çok yaklaşılan ve potansiyeli yüksek olan teknik yakın kızılötesi spektroskopisi tekniğidir. Tez çalışmasında yakın kızıl ötesi spektroskopisi tekniği üzerine yoğunlaşılıp deneysel çalışmalarda bu tekniğine göre yapılmıştır. Yakın kızıl ötesi spektroskopisi tekniği ışığın geçtiği ortamdaki absorbsiyon oranlarını temel alan bir tekniktir. Ölçüm aralıkları 750 nm ile 2500 nm arasındadır. Bu aralıkta kan hücreleri içindeki birçok bileşen ışıkla etkileşime girer. Bu etkileşim, kandaki bileşenlerin dağılımı hakkında bilgi verir. Tez çalışmasında, kan hücrelerinin ana bileşenlerinden biri olan glikoz ayrıntılı olarak incelenmiştir. Glikoz absorbsiyon aralıkları ve ayrıca yakın kızılötesi ışığa karşı glikoz reaksiyonları izlenmiştir. Deneysel çalışmalar üç gönüllü birey üzerinden yürütülmüştür. Deneylerde kullanılmak üzere üç farklı ışık dalga boyunda ışık kaynağı kullanılmıştır. Deneysel çalışmalarda glikoz değişiminin gözlenebilmesi için deneysel kurgu geliştirilmiştir. Bu kurguya göre glikozun kandaki oranının az olduğu açlık durumu, glikoz yükleme durumu 1 ve glikoz yükleme durumu 2 olmak üzere üç farklı ölçüm alınmıştır. Toplamda gönüllüler ile her biri 30 saniye olacak şekilde 27 farklı ölçüm yapılmıştır. Yapılan deneysel çalışmaların sonucuna göre kandaki glikoz miktarı ile ışığın absorbsiyon miktarı değişim göstermektedir. Bu değişimin nonlineer olduğu gözlemlenmiştir. Ölçümlerde, ölçüm platformunda meydana gelen çevresel etkiler ve herhangi bir hareket göz ardı edilmemesi gereken belirgin bir etkiye neden olur. Ayrıca ölçüm probunun daha kararlı hale getirilmesiyle ölçümlerde daha iyi sonuçlar alınabileceği gözlemlenmiştir. Gelecek çalışmalarda ölçüm probuna sıcaklık ve nem sensörleri eklenerek çevresel faktörlerin etkisi azaltılacaktır. Ölçüm probunun kararlı hale getirilmesi için prob içerisinde kullanılacak tüm elektronik bileşenlerin kararlı halde durmasını sağlayacak tasarımlar yapılacaktır.
Comparative analysis of classification techniques for network anomalies management
Today, the rapid development in technology is enabling billions of devices to communicate with each other. This development requires new network technologies to allow all these devices to connect to network easily. In recent years, cyber-attacks have been a serious threat to governments, businesses and individuals. Many Intrusion Detection Systems, which were designed to prevent these cyber-attacks failed. Intrusion Detections Systems (IDS) could not sufficiently recognize the attacks and the cunning ways the attackers used, resulting in inefficient IDS solution and vulnerable networks. It would be a much smarter solution to counteract attacks by using machine learning based systems that is the result of data mining and statistics. This approach will provide a more efficient IDS solution than a conventional IDS solution based on attack recognition techniques. The purpose of this thesis is to propose a method for Network Anomaly Detection System (NADS) using machine learning algorithms with the aim of enhancing the processes of the network troubleshooting, and raising the efficiency of the maintenance processes. This study compares the performance of four selected machine learning classifiers with each other. The selected algorithms are: K-Nearest Neighbors (KNN), K-means, Naïve Bayes and Random Forest. This comparison is conducted to detect the network anomaly and analyze the performance of the classification framework. This comparison is conducted to provide recommendations related to the framework selection. The above mentioned algorithms are implemented and tested on KDD CUP99 intrusion detection dataset that is widely used to evaluate intrusion detection prototypes. The experimental outcomes demonstrate that KNN algorithm perform well in terms of accuracy and computation time. Furthermore the results show that KNN has a successful detection of potential threat of 98.0379 % of all known attacks.
Network monitoring system using machine learning comparative analysis of classification techniques for network traffic monitoring
Online network traffic classification continues to be the focus of long-term interest. Network traffic monitoring and analysis can be done for many different reasons. Generally, it provides raw data input for network monitoring, Quality of Service (QoS) and intrusion detection. Specifically, network traffic monitoring enables the network analyst to understand network resources use and identify network performance. With this information, network analyst may adjust QoS policies to control and manage network resources. This aim is achieved by setting priorities for specific types of data in the network and logging the traffic to comply with the regulations. Network traffic monitoring can be used to create models for academic research. In this thesis, a machine-learning approach that accurately classifies network traffic using Decision Tree Algorithm (DT) is presented and implementing the Principal Component Analysis (PCA) Algorithm for reduction, side by side, to reach the best optimization. Machine learning technology will generate better solutions to monitor and classify network traffic as a result of highly accurate data mining technics and advanced statistics. The purpose of this thesis is to build a Network Monitoring System (NMS) using modern machine learning technologies that works in both online and offline modes. DT algorithm; one of the available data mining algorithms; is used to build the classifier of network. The experiment's results showed that NMS based system has 97.7486 % accuracy (ACC) in successfully classifying the network traffic.
Efficient image annotation and caption system using deep convolutional neural networks
In recent years, with the advances in the artificial intelligence field, image annotation also known as image description (IAC) has progressively attracted researchers' attention. IAC automatically creates natural text descriptions according to the image contents. IAC combines the knowledge of computer vision and natural language processing. In this research, a novel image annotation and description system was developed. The main parts of the developed system are Convolution Neural Network (CNN) and Long Short Time Memory (LSTM). Also, the developed system was enhanced by multiple steps such as adding regularizing to convolution layers, adding dropout layers to the fully connected layers, using genetic algorithms to find the most suitable batch size, and investigating the performance of multiple optimizers such as Adaptive Moment Estimation (Adam), Stochastic Gradient Descent(SGD), and Nesterov accelerated gradient to find the most suitable one for the developed approach. The developed system was validated by multiple experiments using one of the challenging datasets, i.e., the Flicker dataset. Overall, our improved model outperformed the existing state of arts using the BLEU metric. Also, results prove that the designed system can effectively describe images. Last but not least, this research help researchers by highlighting some open challenges in the field of image annotation.
Towards only-vision autonomous wheelchair: A deep learning obstacle detection and image-based avoidance
A rapidly increasing number of people need to use a wheelchair (WC). WC users face several challenges while using the WC. Obstacle avoidance is one of these challenges. Avoidance for some users cannot be done simply using manual control. We propose that the WC should be able to achieve the avoidance automatically. Previous systems offered solutions to similar problems using a fusion of expensive depth sensors. This system uses vision-only technology, via a single camera, to achieve detection and avoidance at a cost that makes it accessible to a large number of disabled users. Our approach integrates functionalities from deep learning, computer vision and mobile robotics fields into the standard powered wheelchair (PWC). A deep-learning model is adapted using learning-transfer techniques to detect obstacles. A dataset of sidewalks has been developed to be used in the learning-transfer process. Any obstacle detected in front of the WC is avoided using a developed image- space avoidance method. The system was deployed during experiments on a Hardware setup using a real PWC. Object detection accuracy is reported as 61% mAP, which is comparable to methods implemented using standard computers. The control module generated the required motor speeds to avoid the obstacle successfully. The overall system achieves a speed of 5 FPS. We conclude that our cost-effective system can work effectively without the need for redundant and costly depth sensors. Adopting our system will increase the mobility of people with disabilities both indoors and outdoors. This opens the way for a vision-only fully autonomous WC.
Yazılım destekli aydınlatma kalitesi ve verimliliği arttırma uygulaması
Günümüzde insanların ortamda bulunan renk sıcaklığı ve aydınlatma düzeyinden psikolojik olarak etkilendikleri, farklı renk ve tasarımdaki ortamlarda bulunan insanların aynı uyaranlara farklı tepkiler verebildikleri bilinmektedir. İnsanların bulunduğu ortamlarda kullanılan aydınlatmanın niteliği büyük önem taşımaktadır. Aydınlatmanın niteliğini belirleyen ana unsurlar, ışık akısının nicel ve nitel özellikleridir. Nicel özellikler; ışık akısı, büyüklüğü ve aydınlık düzeyidir. Nitel özellikler ise, ışığın renk sıcaklığı ve renksel geri verimdir. Bu ana unsurlar dışında parıltı vb. yan unsurlar da mevcuttur.Bu çalışmada; dikkate alınan ortamın renk sıcaklığı ve aydınlatma seviyesi değeri, sensör kullanılarak optimize edilmektedir. Optimizasyonda ölçülen değerler ile veri kümeleri oluşturularak, verimli ve ekonomik tarzda ideal aydınlatmaya yaklaşım hedeflenmektedir. Bu veri kümelerinden yararlanılarak, PIC mikrodenetleyici ve PLC birleşik sistemiyle, renksel ışık ve aydınlatma düzeylerinin, istenilen ayar değerlerine yaklaştırılması amaçlanmaktadır.Bu tez çalışmasında, kullanılan sensörün aydınlatma ortamında bulunduğu pozisyona bağlı olarak aydınlatma kalitesi ve verimdeki değişimin gözlemlenmesi amacıyla, farklı noktalarda ölçümler yapılmıştır. Yapılan ölçümler analiz edilerek, en uygun sensör pozisyonu tespit edilmiştir. Ortam aydınlatması için kullanılan set değerlerinin aydınlatma parametrelerine göre tanımlamaları yapılmış ve set değerlerinin daha anlaşılır olmaları sağlanmıştır. PIC mikrodenetleyici kullanılarak tasarlanan analog sayısal dönüşümü kartları ile hem set değeri, hem de sensörden okunan analog gerilim değeri, sayısal veri formuna sokulmuştur. Bu sayısal veriler, PLC cihazı kullanılarak karşılaştırılmış ve her iki veri arasındaki fark, uygulamada kullanılan ışık kaynakları ile telafi edilmiştir. Sensörün gün ışığından da bir gerilim değeri okuması sayesinde enerji tasarrufu sağlanmış olup, ölçülebilir tasarruf oranları elde edilmiştir.Anahtar KelimelerPLC (Programmable Logic Controller) , PIC16F877, aydınlatma otomasyonu, yazılım, renk, ışık
MR taramaları üzerinde derin öğrenme modelleri kullanılarak multipl skleroz lezyonlarının otomatik tespiti ve değişiminin takibi
Multipl Skleroz (MS) hastalığı genç ve orta yaşlı insanlarda sıkça görülen beyin ve omuriliği etkileyen, yaygın bir merkezi sinir sistemi (MSS) rahatsızlığıdır. MSS içinde bulunan sinir ağları üzerindeki kılıf yapısında oluşan iltihap sonucunda meydana gelen hastalık kişide önemli bilişsel kayıplara neden olur. MS hastalarında fonksiyonel yeti kayıpları ve ileri safhalarda geri dönülemez beyin hasarları da görülebilir. Bu nedenlerden dolayı MS hastalığının erken aşamada tespiti ve takibi oldukça önemlidir. MS'in teşhisinde ve takibinde manyetik rezonans (MR) görüntüleme yaygın olarak olup teşhis için genellikle en az iki farklı periyottaki taramalarda MS lezyonlarının görülmesi beklenmektedir. Hastalığın takibi MR kesitlerindeki lezyonların değişimlerinin izlenmesi ile yapılmaktadır. MS lezyonlarının boyutlarının küçük olması ve beyindeki başka nörolojik rahatsızlıklara benzemesi, MS lezyonlarının tespitini zorlaştırmaktadır. Bu tez çalışmasında, MR görüntüleri üzerinde MS lezyonlarının otomatik tespiti için iyileştirilmiş bir derin öğrenme modeli (iMask R-CNN) önerilmiştir. Bu modelde Maske-Tabanlı Bölgesel Evrişimsel Sinir Ağı (Mask R-CNN) mimarisinin tahmini önemli bölge hizalama (RoIAlign) işleminde iyileştirme yapılmıştır. Ayrıca, MS lezyonlarının tespiti için 52 hastadan periyodik olarak alınan FLAIR MR kesitlerini içeren MSAkdeniz isimli bir MR veriseti hazırlanmıştır. Tez kapsamında, ek olarak, MS lezyonlarının otomatik tespiti ve değişimin takibi için, hekimlere yardımcı olabilecek, iMask R-CNN modelini kullanan DeepMSWeb isimli web tabanlı bir karar destek sistemi (KDS) uygulaması geliştirilmiştir. Bunun yanında, MS lezyonlarının otomatik tespiti için, önerilen iMask R-CNN modeli başta olmak üzere, MSAkdeniz veriseti ve eHealth, UMCL, ISBI2015 ve MICCAI2008 verisetleri üzerinde, iki faklı platformda beş farklı derin öğrenme modeli kullanılarak deneysel çalışmalar yürütülmüştür ve sonuçlar karşılaştırılmıştır. Önerilen iMask R-CNN kullanılarak, lezyon tespitinde, verisetlerinde Dice benzerlik katsayısı (DSC) için %88.32±4.44 skorları, lezyon doğru pozitif oranı (LTPR) için %89.80±3.54 skorları elde edilmiştir. Önerilen iMask R-CNN ve DeepMSWeb yapıları ile MS lezyonlarının tespiti ve değişiminin takibinde kullanılabilecek yardımcı araçlar geliştirilmiştir.
Azami genlik algılı yüksek çözünürlüklü zaman-frekans analizi
Çalışmada, sayısal ses işaretlerinin incelenmelerinde sıklıkla kullanılan literatürde mevcut farklı tür zaman-frekans analizleri incelenmiş ve kısa zamanlı Fourier dönüşümü temel alınarak yüksek zaman ve frekans çözünürlüklü yeni bir zaman frekans analiz yöntemi geliştirilmiştir. Fourier dönüşümündeki zaman belirsizliği problemine karşılık kısa zamanlı Fourier dönüşümü (KZFD) yöntemi ortaya çıkmıştır. KZFD'de aynı anda hem yüksek zaman çözünürlüğü hem de yüksek frekans çözünürlüğü elde etmek mümkün olamamaktadır. Buradan yola çıkılarak, bu çalışmada, KZFD'nin farklı zaman pencerelerindeki üstünlükleri kullanılarak sentezleme yöntemiyle aynı anda yüksek zaman ve frekans çözünürlüğü sunan Azami Genlik Algılı Yüksek Çözünürlüklü Zaman - Frekans Analizi (AGA–ZFA) yöntemi önerilmiş ve MATLAB® kullanılarak geliştirilen program ile analiz işlemleri gerçeklenmiştir. Geliştirilen yöntemin başarımının nicel olarak ölçümü gerçekleştirilerek AGA–ZFA yöntemi, literatürde mevcut ZFD yöntemleriyle karşılaştırılmıştır. Sayısal sonuçlar zaman frekans analizinde, geliştirilen yöntemin diğer yöntemlere oranla %52'ye kadar iyileştirme sağladığını göstermiştir.
Düşük iletişim yüklü olay-tetikli dağıtık kestirim
We propose a novel algorithm for distributed processing applications constrained by the available communication resources using diffusion strategies that achieves up to a 10^3 fold reduction in the communication load over the network, while delivering a comparable performance with respect to the state of the art. After the computation of the local estimates, the information is diffused among the processing elements (or nodes) non-uniformly in time by conditioning the information transfer on level-crossings of the diffused parameter, resulting in a greatly reduced communication requirement. We provide the mean and mean-square stability analyses of the proposed algorithm, and illustrate the gain in communication efficiency compared to other reduced-communication distributed estimation schemes.
Radar anten taramalarının otomatik analizi
Estimation of the radar antenna scan period and recognition of the antenna scantype is usually performed by human operators in the Electronic Warfare (EW)world. In this thesis, we propose a robust algorithm to automatize these two criticalprocesses. The proposed algorithm consists of two main parts: antenna scanperiod estimation and antenna scan type classification. The first part of the algorithminvolves estimating the period of the signal using a time-domain approach.After this step, the signal is warped to a single vector with predetermined size(N) by resampling the data according to its period. This process ensures thatthe extracted features are reliable and are solely the result of the different scantypes, since the effect of the different periods in the signal is removed. Four differentfeatures are extracted from the signal vector with an understanding of thephenomena behind the received signals. These features are used to train naiveBayes classifiers, decision-tree classifiers, artificial neural networks, and supportvector machines. We have developed an Antenna Scan Pattern Simulator (ASPS)that simulates the position of the antenna beam with respect to time and generatessynthetic data. These classifiers are trained and tested with the syntheticdata and are compared by their confusion matrices, correct classification rates,robustness to noise, and computational complexity. The effect of the value of Nand different signal-to-noise ratios (SNRs) on correct classification performanceis investigated for each classifier. Decision-tree classifier is found to be the mostsuitable classifier because of its high classification rate, robustness to noise, andcomputational ease. Real data acquired by ASELSAN Inc. is also used to validatethe algorithm. The results of the real data indicate that the algorithm isready for deployment in the field and is capable of being robust against practicalcomplications.
İnsan gözünün topoğrafik görüntülerının özelliklerini elde etmek için akilli algoritma oluşturmak
Precise diagnosis for a wide range of diseases infecting the human eye is a commitment. Therefore, developing new, smart algorithms is necessary to enhance doctors' diagnostic decisions. The recently invented Pentacam® is a measurement system that introduces topographic maps for the cornea, measures changes upon it, and helps doctors to make a precise diagnosis. This study extracts features from corneal topographic maps to improve the Pentacam® readings and further support precise diagnosing by using deep learning techniques, with an analytical view of the extracted features. A 16-layer convolutional neural network (CNN) was trained using the VGG-16 network to extract powerful features from corneal topographic maps. A sample of 732 human eyes were selected from enlarged topographic images from both genders (414 females and 318 males), divided into two groups: normal and abnormal. The patients' ages ranged from 12 to 76 years. The procedure of the study consisted of three major steps: (1) classification of the extracted features according to the refractive map type (where the estimated accuracy was 96.6%); (2) prediction of the clinical state (normal or abnormal) per individual map (where the estimated accuracy was 88.8%, 98.9%, 94.8%, and 94.5% for the sagittal map, the elevation front map, the elevation back map, and the corneal thickness map, respectively); and (3) comparison of the predicted results and clinical decision-making. The agreement between them reaches about 94.72%, which indicated the power and usefulness of the proposed algorithm.
İç ortamlarda insan-robot etkileşimi
Bu tez çalışmasında robotik alanında tasarlanan İnsan-Robot Etkileşimine (İRE) diyalog kurma ve geliştirme konusunda etkili çözümler üreten üç model önerilmiştir. Robotların iç ortamlarda insanlarla birlikte çalışması veya insanları bilgilendirmesi amacıyla kullanılmak istendiğinde karşılaşılan ilk problemin insan robot arasındaki diyaloğun nasıl geliştirilmesi gerektiği ve etkili diyaloğun nasıl sağlanması gerektiğidir. Bu amaçla İRE alanındaki çalışmalar incelenmiş ve kullanılan yöntemler araştırılmıştır. Bu çalışmalardan esinlenilerek, İRE alanında, robotik sistemlerde kullanılabilecek diyalog modelleri üzerinde geliştirmeler yapılmıştır. Farklı diyalog modelleri için sonlu durum makinesi, içerikten bağımsız gramer yöntemlerinden faydalanılmıştır. Bu kapsamda sonlu durum makineleri ve içerikten bağımsız gramer yöntemleri incelenmiştir. Bu yöntemlerle diyalog modellemeleri geliştirilmiştir. Gerçekleştirilen bu diyalog modelleri gerçek zamanlı olarak robot üzerinde denenmiştir. Önerilen modellerde görüntü işleme ve ses işleme algoritmaları ile insan robot arasında interaktif etkileşimi sağlayan araçlar kullanılmıştır. İRE ile alakalı iki adet deney ortamı oluşturulmuştur. İlk deneyde robot insanı bulduğu zaman kafasıyla karşısındaki insanının yüz hareketlerini taklit ederek kafa takibini başlatıyor. Robot başlangıç durumunda karşısındaki insandan etkileşim beklemektedir. Eğer robotun karşısındaki insan robotla belirlenen süre içerisinde etkileşime geçmez ise robot karşısındaki insana kendisini nasıl kullanmasını anlatarak etkileşimi başlatıyor. İnsan etkileşime geçer ise robot sorulan soruları yanıtlamaktadır. Etkileşim robotu programlayan programcının belirlediği Sonlu Durum Makinesine (SDM) göre tamamlanıyor. Üç farklı SDM önerilmiştir. Bunlar "İnsan kontrol modeli", "Kafa takip modeli" ve "Robot gözünden kontrol modeli" olarak adlandırılmaktadır. Oluşturulan iki deney ortamı önerilen üç model ile çaprazlanarak altı deney beş tekrar ile gerçekleştirilmiştir. Yapılan deneylerde SDM modelleri arasında başarılı etkileşimler belirlenmiş. En uygun model tespit edilmiştir. İleride yapılacak çalışmalarda insanların yaş, cinsiyet ve duygu analizi yapılarak bu oranın daha da arttırılması hedeflenmektedir.
Beyin tümörü tedavisi için kemoterapi ilaç salınım sistemi modellenmesi ve kontrolü
Beyin tümörleri, merkezi sinir sisteminin karmaşık yapısı nedeniyle erken tanı ve etkili tedavi süreçlerini zorlaştıran, yüksek mortaliteye sahip kanser türlerinden biridir. Tedavi sürecinde en sık kullanılan yöntemlerden biri olan kemoterapi, sağlıklı hücreleri de etkileyebilmekte ve hastanın bağışıklık sistemini baskılayabilmektedir. Bu nedenle kemoterapide uygulanan ilaç dozunun bireyselleştirilmiş, güvenli ve optimum şekilde kontrol edilmesi büyük önem taşımaktadır. Bu çalışmada, beyin tümörü tedavisinde kemoterapi ilaç dozajının kontrolü için Tip-2 Bulanık Mantık Denetleyici (T2BMD) tasarlanmıştır. T2BMD'nin, üyelik fonksiyonlarındaki belirsizlikleri daha başarılı bir şekilde yönetmesi sayesinde daha esnek ve hassas kontrol mekanizmaları sunduğu gösterilmiştir. Denetleyici sisteme ait parametreler doğadan esinlenen yeni nesil bir algoritma olan Yapay Sinekkuşu Algoritması (Artificial Hummingbird Algorithm, AHA) ve literatürde oldukça yaygın olan Parçacık Sürü Optimizasyonu (PSO) algoritması ile optimize edilmiştir.
Derin öğrenme kullanilarak cilt kanserinin tespiti ve siniflandirilmasi
One of the worst malignancies is skin cancer. It is expected to spread to other parts of the body if it is not identified and treated at the outset. The method for successfully treating skin cancer uses both image processing and deep learning. In this research, three different kinds of skin cancer, including melanoma, pigmented benign keratoses, and basal cell carcinoma are introduced for detection and classification using efficient methods in Machine Learning (ML) such as K-Mean clustering and Multi-class support vector machine (M-SVM) algorithm, and then Deep Learning (DL) techniques such as AlexNet and ResNet are used. Additionally, deep convolutional neural networks (CNN) effectiveness and capacity are seen. The data set contains 1176 images of skin cancer with different classes of disease, this data set is used in both ML and DL. In ML the data set is divided into training and testing sets, these sets are pass through a few steps of enhancement using a canny edge detection filter and extracting the features using the Gray-Level Co-occurrence Matrix (GLCM) method. Then, these images are segmented into three clustering using K-mean clustering algorithms. In DL two models are used AlexNet and ResNet, in these models the data sets are divided into training and testing sets. Then, an augmentation technique has been proposed, it's very useful for small data sets, and the results show changes in accuracy result. The results show an accuracy of 99.03%, and 97.02% for AlexNet, and ResNet respectively.
Elektrik akıllı şebeke stabilitesi için yapay zeka modellerinin uygulanması
This study does a comparative analysis of several models, including XGBoost, SVM, Random Forest, KNN, Logistic Regression, Decision Tree, Neural Network, SimpleRNN, LSTM, and GRU. The performance of these models is evaluated based on metrics such as accuracy, precision, recall, and F1-score. The XGBoost Classifier is considered the optimal model due to its superior combination of precision, computational effectiveness, and interpretability. The findings presented in this study underscore the considerable capacity of machine learning in facilitating predictive analytics within the context of smart grids. Moreover, these results establish a robust basis for further research endeavours in this domain. Nevertheless, the literature highlights some challenges, such as the intricate nature of models, their limited interpretability, and the substantial computational demands they impose. These issues underscore the necessity for more research and enhancements in this field. As the study concludes, this paper offers strategic recommendations for effectively integrating the findings into actual applications of smart grids. Additionally, it outlines potential avenues for future research in this field.
Kullanilan iot cihazlarinda kullanici gizliliğimakine ve derin öğrenme yaklaşimlar
The swift expansion of Internet of Things (IoT) technology has sparked concerns regarding the privacy of users, since these devices often collect and transmit vast amounts of personal information. To address these issues, this thesis will look at the use of machine and deep learning technologies to improve user privacy on IoT devices. First, the research will examine the existing state of user privacy on IoT devices, as well as the issues associated with protecting privacy. This will include a discussion of the many types of data gathered and communicated by IoT devices, as well as the numerous ways in which this data might be exploited or hacked. The research will also look at current legislative frameworks and best practices in the sector for preserving user privacy on IoT devices. The thesis will then investigate the application of machine learning approaches to improve user privacy on IoT devices. This will take a look at the many machine learning techniques that may be used for this, such as decision tree algorithms and ANNs. The research will also look at the possible benefits and drawbacks of utilizing these algorithms for privacy protection, such as the trade-offs between privacy and other objectives like performance or accuracy. Besides machine learning, the project will look into the use of deep learning technologies for improving user privacy on IoT devices. Deep learning models, a specific category of machine learning techniques, have demonstrated significant potential across various applications. The research will examine the potential benefits and challenges of applying deep learning algorithms for privacy protection on IoT devices, as well as the present related works in this field. Finally, the dissertation will conclude with a discussion of the potential future direction of research in this area, including the potential for integrating machine and deep learning approaches with other privacy-enhancing technologies and the potential for additional regulatory or industry-led efforts to improve user privacy on IoT devices. This dissertation intends to offer a complete assessment of the present status of user privacy on IoT devices, as well as the possibilities for enhancing privacy via the application of machine and deep learning technologies. The study intends to contribute to continuing efforts to secure user privacy in the rapidly developing realm of IoT by addressing these challenges.
Değiştirilmiş SVM iş akışı kullanılarak meme kanserinin yerleştirilmesi ve tespiti
Mammography is the most effective method in the early detection of breast cancer, which can detect up to 90% of cases. mammography was used in only 24% of cases in 2017, which is below the 70% expected by the World Health Organization (WHO). This explains one of the causes of late diagnosis and the increase in mortality. According to WHO data, the average time to start treatment is 120 days after the first appointments. This situation is due to the delay in scheduling consultations, in addition to the fact that many times excessive tests are requested, which slow down the diagnostic process. In this paper, we Propose a medical decision support model in breast cancer to help diagnose more quickly, using prototype selection associated with (Support Vector Classifier (SVC), K-Nearest Neighbor (KNN), Random Forest (RF), Logistic Regression (LR)). In this way, we expect to obtain robust results with agility and efficiency.
Eliptik eğri uygulaması nesnelerin interneti için kriptografi(IOT) güvenlik
The Internet of Things, or IoT, is essential to the fields of industry, healthcare, and information technology, among others. It is made up of numerous interconnected things that are in communication with one another. Because approved objects, in addition to users, may access data regarding the Internet of Things. For IoT applications and technology to be widely adopted, security is a necessary component. To improve the security of the IoT data, this study suggests an Elliptic Curve Cryptography approach. Two keys are used in Elliptic Curve Cryptography (ECC): a public key and a private key. The user uses the private key for encryption, and the public key is used for user identification during authentication. Similar to this, using the private key, the sender encrypts; if secrecy is desired, using the public key, one can decrypt the communication. Selecting the private key is a problem with every public key. Random selection of small values raises concerns about the overall algorithm's security. considering the values. This study suggests using the Cuckoo Search Algorithm to select values at random.
MFF-LSTM: Çok Ölçekli Özellik Füzyon Tabanlı Uzun Kısa TasarlamaYalan Haber Tespit Sistemi için Farklı Özelliklere Sahip Dönem Belleği
The rise in the usage of media has led to a significant increase in the raise of false information, making it imperative to combat this issue and reduce our reliance on such unreliable sources. Fake news can mislead people, spread rumors, and even impact the positions of political leaders. Detecting fake news has become crucial in this digital era, with direct messaging platforms and social media playing a major role in its proliferation. Various innovative techniques have been suggested to determine fake news, making the endeavor both intriguing and challenging. Hence, this synopsis aims to develop the adaptive learning model with multiscale feature fusion for fake news detection. The proposed system constitutes "text collection, text pre-processing, feature extraction and detection". Initially, the text input is collected from the benchmark datasets, which is then followed by the text stage of pre-processing. Here, the pre-processed text is obtained that is fed into the feature extraction phases. The three feature extraction techniques such as "Bidirectional Encoder Representations from Transformers (BERT), Term Frequency-Inverse Document Frequency (TF-IDF) and GloVe Embedding" are employed to provide the feature set 1, 2 and 3. Finally, these resultant features are given to "Multiscale Feature Fusion based Long Short Term Memory (MFF-LSTM)", where the features are fused together in multiscale manner and detection is taken place by LSTM. Therefore, the system evaluation is done by considering the distinct measures and compared among traditional approaches. Hence, the recommended model attains the desired results to detect the fake news that helps to evade the exploration of any false information.
Makine öğrenmeyi kullanarak COVID-19 aşı görüşlerinin duyarlılık analizi için sınıflandırma algoritmalarının karşılaştırmalı bir çalışması
The task of analyzing textual data and classifying them into positive, negative, or neutral emotions within the domain of natural language processing is a multifaceted undertaking. The primary objective of this study is to employ machine learning algorithms in order to classify opinions pertaining to the coronavirus disease and vaccines. In this study, four algorithms were employed, namely Random Forest (RF), Gradient Boosting Classifier (GBC), Logistic Regression (LR), and Decision Tree (DT). The RF and GBC algorithms demonstrated a commendable accuracy rate of 89%, while the LR and DT algorithms yielded a slightly lower accuracy rate of 87%. The findings derived from this research can provide valuable guidance to policymakers in effectively addressing potential barriers that may impede the successful execution of vaccination campaigns. The analysis of the Kaggle data, which encompasses a wide range of commentaries related to the pandemic and vaccines, underscores the urgent need for prompt measures to attain herd immunity against Covid-19. This imperative objective holds significant importance in effectively managing the transmission of the virus and mitigating its adverse consequences on the well-being of the general population. The task at hand necessitates the acknowledgment and resolution of public apprehensions, as well as the establishment of trust and assurance in the vaccination initiative. This study presents an analysis of the machine learning techniques employed and conducts a comparative evaluation of their significance. The forthcoming research endeavors to create an application that will be capable of categorizing sentiments and opinions pertaining to diseases and vaccines. It is imperative for governments and organizations to comprehend the obstacles linked to the worldwide COVID-19 vaccination endeavor in order to develop efficacious strategies. Nevertheless, it is crucial to acknowledge that the scope of the analysis was restricted to tweets written in the English language. This limitation may potentially undermine the credibility and generalizability of the findings pertaining to overall sentiment. Additional investigation could be conducted to examine more extensive Twitter datasets employing deep learning models in order to gain a deeper comprehension of the general public's attitudes towards COVID-19 vaccines.
5G iletişim sistemi için yeniden yapılandırılabılır akıllı meta yüzey
For 5G mobile communication networks, a novel antenna array is built using a printed monopole based on a metamaterial (MTM). For 5G communication networks, the architecture and functionalities of a Multiple-Input Multiple-Output (MIMO) antenna array are anticipated. Four components make up the MIMO antennae array, which is stacked in an octagon 3D array arrangement consecutively. The components are arranged to create a particular pattern. The array is constructed on a 25 x 25 x 30 mm³ FR4 substrate. Because of its RF characteristics, this substrate option is frequently used in antenna designs. A split ring resonator (SRR) is connected to each antennae element, which is a printed circuit monopole, via a T-resonator (TR). Resonances at 3.6 GHz and 4.8 GHz are provided by this design, making it appropriate for sub-6GHz 5G networks. To lessen mutual coupling, electromagnetic band gap (EBG) spacers are utilised between antenna elements. The antennae array's concerts are improved by the EBG spacers' moonquake fractal-shaped array design. The moonquake fractal EBG spacers are placed to achieve a -20dB mutual coupling reduction between antennae elements. This reduction in mutual coupling is important for maintaining good isolation and minimizing interference between elements. Four PIN diodes are connected to each antennae element to control the antennae's concerts . By using these diodes, different switching scenarios are tested to understand their impact on antennae behaviour. The projected array's
Mülteci ve göçmenlerin etkili tespiti Türkiye'nin çevresinde insanlar evrimsal sinir kullanıyor ağ
In the years following the outbreak of the Syrian Civil War in 2011, non-governmental organizations (NGOs) played a crucial role in assisting refugees and distributing relief to people in need so that
Derin özelliklere sahip enerji verimli bir SDN-ıot mimarisi IoT destekli akıllı şehir için öğrenme tabanlı trafik tahmini
Thesis proposes an energy-efficient SDN-IoT architecture tailored for IoT-enabled smart cities. This architecture addresses the challenges of resource optimization and energy consumption management in the context of diverse IoT devices and dynamic traffic patterns. The architecture provides a framework for efficient network management and facilitates sustainable operation in smart city environments. A modified SVM (Support Vector Machine) algorithm is introduced and integrated into the proposed architecture for traffic prediction. By enhancing the traditional SVM algorithm with specific modifications tailored for IoT-generated data, the thesis contributes to improving the accuracy and reliability of traffic prediction in smart city networks. The modified SVM algorithm can effectively capture complex patterns and variations, enabling precise anticipation of traffic demands. The research presents a comprehensive evaluation of the proposed architecture's performance. Through extensive simulations and practical deployments in a smart city testbed environment, the paper assesses the energy efficiency, resource utilization, and predictive accuracy achieved by the architecture
Koronavirüs (COVİD-19) sürecinde 5g sağlık sistemi
This study's primary goal is to identify knowledge gaps and the most often used information mining techniques in the body of current literature. Novel coronavirus pneumonia has become a significant public health concern, particularly COVID-19. In the academic community, information extraction techniques have been used to uncover concealed knowledge as the possibility of a widespread raises open concerns for wellness. A methodical investigation was carried out using the use of the PubMed, Scopus, and Internet of Science databases. There were 515 million COVID-19 cases and about 6 million deaths worldwide as of May 2022. According to estimates from the World Health Organization, 115,000 healthcare workers died from COVID-19 between January 2020 and May 2021. Patients with impaired immune systems and nano-oncology for the Web of Nano Technologies (IoNT).. There are still issues that need to be clarified and addressed in order to plan clinical execution. These cover the following topics: foundation and breadth; health risks; security and comprehension of information security; artificial intelligence; blockchain; IoT and 5G utilization; authorization; ongoing acknowledgement; and final client education on these technologies.
Dünyadaki siber suçları azaltmak için gelişmiş yapay tabanlı siber güvenlik ağı
In the current digital epoch, cybersecurity emerges as a paramount concern, given the relentless advancement of cyber threats that pose formidable challenges to global digital infrastructures. This thesis embarks on a comprehensive exploration, centering on advanced neural network models, particularly emphasizing the Convolutional Neural Network (CNN), to address the imperative need for resilient cyber defense mechanisms. Employing meticulous experimentation and analysis utilizing the 'Cyber Security Indexes' dataset, this study meticulously evaluates the performance of these models across a spectrum of cyber-attack types. The findings illuminate the CNN model's robustness and efficacy, portraying its potential in fortifying cybersecurity measures and countering the evolving landscape of threats. Throughout this exploration, with a focus on achieving a 96% accuracy threshold, the study outlines the implications of these advancements in fostering a more secure digital landscape on a global scale. The comprehensive insights drawn from this research collectively underscore the pivotal role of the CNN model in fortifying cybersecurity defenses, offering a beacon of hope in mitigating the escalating cyber threats prevalent in today's digital milieu.
Bir roman dört yönlü DNN modeli için siber saldırı tespiti ve önlenmesi
In an era marked by escalating cyber threats, traditional security measures struggle to contend with the surging prevalence of cyber-attacks. To address this challenge, we present a groundbreaking solution in the form of the Quad Directional Recurrent Neural Network (Quad-RNN). This novel architecture, featuring four directions of input and output, amalgamates the strengths of Bidirectional RNNs (BRNNs) and Simple RNNs. Our evaluation, conducted on the NSL-KDD and DDoS datasets, establishes the superiority of Quad-RNN over BRNN and Simple RNN counterparts. Demonstrating enhanced accuracy, precision, recall, and F1 score, the Quad-RNN architecture notably diminishes false positives. This research heralds a pivotal advancement in the realm of cyber-attack detection and prevention, addressing the imperative for resilient and adaptive security measures in the face of evolving threats.
Göğüs röntgeni görüntülerinde çocuk pnömonisinin otomatik olarak tespitinde derin öğrenme ve makine öğrenme yöntemlerinin etkinliğinin değerlendirilmesi
Accurate identification and classification of images of patients with pneumonia is vital for effective diagnosis and treatment. Advanced learning techniques such as CNN LSTM (Convolutional Neural Network Long Short-Term Memory) have proven 97% accuracy on this task. A labeled dataset of 5,856 images of pediatric pneumonia patients from Kaggle was used to evaluate these models. The dataset was divided into training (70%), testing (15%), and validation (15%) sets. It is worth noting that the ResNet 50 -CNN model and the CNN-ExtraTrees model achieved an accuracy rate of 98% and 94%, respectively. These models provide resources for clinicians and radiologists in their decision-making processes. Extracting features from images plays a role in pattern prediction using the optimizer as part of the modeling process. To improve performance, modifications were made to the RMSprop parameters. Model performance evaluation included metrics such as recall, precision, F1 score, and overall accuracy evaluation. These measures provide insight into the effectiveness of models while also serving as reference points for investigations. The exceptional accuracy rates achieved by these models confirm their ability to accurately classify patients with pneumonia, which has implications for improving patient outcomes. The evaluation criteria used in this research provide an examination of the advantages and limitations of models that highlights the importance of incorporating these metrics to ensure accurate performance in scenarios. Keywords: Medical Image, Pediatric Pneumonia, Resnet50-CNN, CNN-Extratrees, Convolutional Neural Network, Long Short-Term Memory
Ekonomik analiz ile petrol rafinerisi iş akışı dönüşümü için modern web tabanlı karar destek sistemi
The refinery industry operates within a dynamic and competitive environment, requiring decision-makers to navigate complex challenges and uncertainties. This project focused on the development of a sophisticated Decision Support System (DSS) using advanced technologies such as NET 6 and C#. The primary objective was to empower decision-makers in the refinery industry with a robust tool to analyze extensive datasets, gain valuable insights, and make informed strategic choices. The project followed a systematic development process, involving the transformation of traditional Excel-based models into an automated application. By harnessing the capabilities of NET 6 and C#, the DSS facilitated seamless integration with backend systems, efficient data processing, and complex calculations. The frontend of the DSS was built using ReactJS, resulting in a user-friendly and visually appealing interface. Decision-makers can effortlessly navigate through the system, visualize data, and analyze key metrics, enabling them to make data-driven decisions with confidence. The implementation of the DSS offers significant advantages to the refinery industry. It reduces dependency on manual processes, automates data analysis, and provides real-time access to critical information. Decision-makers can explore various scenarios, evaluate financial implications, and optimize resource allocation. By streamlining the decision-making process, the DSS saves time, enhances operational efficiency, and empowers decision-makers to stay ahead in the competitive refinery industry. This project vii underscores the transformative potential of advanced technologies in leveraging data for informed decision-making, fostering a competitive edge in the refinery industry landscape. Keywords: Oil, Decision Support System, Web-Based Decision, Economic Analysis, Decision Support System (DSS).
Makine öğrenmesi yöntemleri ile çevrimiçi kredi kartı işlemlerinde Fraud analizi tahmini
Kredi kartları, dünya genelinde yaygın kullanımı ve sağlam altyapısı sayesinde hızla insanların günlük hayatlarına entegre olmuş ve güvenle kullanılan ödeme araçlarından biri haline gelmiştir. Ancak, kredi kartı sayılarının artması ve işlem hacminin hızla büyümesi, dolandırıcıları cezbetmiş ve haksız kazanç elde etme amacıyla çeşitli dolandırıcılık yöntemlerini ortaya çıkarmıştır. Günümüzde kredi kartı bilgilerine ulaşmanın kolaylaşması, kredi kartı dolandırıcılarının faaliyetlerini kolaylaştırmaktadır. Gelişen teknoloji ile hesap hareketleri zaman içinde analiz edilebilmekte ve kötü niyetli verilerin kullanımı izlenebilmektedir. Bu çalışmada, Kaggle veritabanından elde edilen Kredi Kartı Dolandırıcılık Teşhis veri seti kullanılarak bu çalışma, kredi kartı dolandırıcılığı tespiti için topluluk tabanlı XGBoost modeli ile diğer geleneksel makine öğrenmesi modellerini karşılaştırarak önemli bulgular ortaya koymaktadır. XGBoost, Random Forest ve CatBoost modellerinin, kesinlik, geri çağırma gibi performans ölçütleri üzerinde daha iyi performans sergilediği belirtilmektedir. Bu sonuçlar, finansal kurumların günlük operasyonlarında karşılaştıkları dolandırıcılık risklerini azaltma potansiyeline işaret etmektedir. Çalışmanın dikkate değer bir noktası, XGBoost, Random Forest ve CatBoost'un dengesiz veri setleriyle başa çıkma yeteneğinin vurgulanmasıdır. Geleneksel modellerin zayıf performans gösterdiği bu durumlarda, XGBoost, RandomForest ve CatBoost'un %99'a kadar tahmin doğruluğu sağladığı ve diğer modellere göre daha iyi bir performans sunduğu gözlemlenmektedir. Anahtar Kelimeler: Çok Katmanlı Sinir Ağları, Veri Madenciliği, Naive Bayes Yönetmi, Sahtekarlık tespiti.