Bilgisayar Mühendisliği Bölümü / Department of Computer Engineering
Permanent URI for this collectionhttps://hdl.handle.net/11413/6817
Browse
Recent Submissions
Publication Restricted Load Frequency Control in Smart Grids: A Review of Recent Developments(Elsevier Ltd., 2024) Wadi, Mohammed; Shobole, Abdulfetah; ELMASRY, WİSAM; Küçük, İsmailThis study provides a comprehensive and fresh review of load frequency control (LFC) in multi-area interconnected power systems (MAIPSs). The central tasks of LFC are to keep frequency variations as minimum as possible to achieve an acceptable level of stability. This research provides a complete view, from early classical control to recent technologies and modern techniques considering strategies, robust, optimal, self-tuning, and adaptive controllers for LFC in MAIPSs. Fuzzy control and earlier and recent optimization algorithms also are analyzed. The linearity, nonlinearity, and uncertainty of LFC models are also investigated. This review emphasizes recent technological advances and novel control strategies. LFC is also considered with the integration of wind, photovoltaic, electric vehicles, and storage devices. Besides, the utilization of machine learning and reinforcement techniques is examined. Further, LFC in smart grids and modern complex power systems concerning limited communication bandwidth, communication failure, and cyber-attacks are also investigated. This review provides an in-depth and detailed diagnosis of the challenges associated with LFC in modern and complex power systems. This work may be valuable for studies and practitioners interested in LFC. It, in detail, investigates future efforts and directions to enhance LFC performance, stability, and reliability in the face of increasing complexity and uncertainty. © 2023 Elsevier LtdPublication Metadata only Turkish Sign Language Recognition Using a Fine-Tuned Pretrained Model(Springer Science and Business Media Deutschland GmbH, 2024) ÖZGÜL, GİZEM; DERDİYOK, ŞEYMA; AKBULUT, FATMA PATLARMany members of society rely on sign language because it provides them with an alternative means of communication. Hand shape, motion profile, and the relative positioning of the hand, face, and other body components all contribute to the uniqueness of each sign throughout sign languages. Therefore, the field of computer vision dedicated to the study of visual sign language identification is a particularly challenging one. In recent years, many models have been suggested by various researchers, with deep learning approaches greatly improving upon them. In this study, we employ a fine-tuned CNN that has been presented for sign language recognition based on visual input, and it was trained using a dataset that included 2062 images. When it comes to sign language recognition, it might be difficult to achieve the levels of high accuracy that are sought when using systems that are based on machine learning. This is due to the fact that there are not enough datasets that have been annotated. Therefore, the goal of the study is to improve the performance of the model by transferring knowledge. In the dataset that was utilized for the research, there are images of 10 different numbers ranging from 0 to 9, and as a result of the testing, the sign was detected with a level of accuracy that was equal to 98% using the VGG16 pre-trained model. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.Publication Metadata only Enhancing Targeting in CRM Campaigns Through Explainable AI(Springer Science and Business Media Deutschland GmbH, 2024) Ayaz, Teoman Berkay; Özara, Muhammet Furkan; Sezer, Emrah; Çelik, Ahmet Erkan; AKBULUT, AKHANModern customer relationship management (CRM) solutions are vital to firms because they streamline the administration of customer interactions, sales processes, and marketing initiatives. To fully exploit the potential of massive volumes of customer data, these platforms need help from AI techniques to quickly evaluate and extract useful insights, personalize customer experiences, and optimize decision-making to improve business outcomes. This study delves into the use of explainable AI methods like SHAP, LIME, and ELI5 to analyze CRM campaign outcomes. The purpose of this research is to discover essential traits that serve as indications for successful targeting by analyzing a dataset that captures the results of customers’ interactions with campaign content as responder or non-responder. Using these methods improves interpretability and closes the gap between AI-driven decision-making and human understanding. The findings add to the field by offering clear rationales for consumer actions, which in turn helps companies fine-tune their targeting tactics and boost the efficiency of their campaigns. This study emphasizes the value of AI systems being transparent and interpretable in order to promote trust and enable data-driven decision-making in CRM contexts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Publication Open Access Blockchain-Based KYC Model for Credit Allocation in Banking(IEEE-Inst Electrical Electronics Engineers Inc., 2024) Karadağ, Bulut; Zaim, A. Halim; AKBULUT, AKHANThe implementation of the Know Your Customer (KYC) strategy by banks within the financial sector enhances the operational efficiency of such establishments. The data gathered from the client during the KYC procedure may be applied to deter possible fraudulent activities, money laundering, and other criminal undertakings. The majority of financial institutions implement their own KYC procedures. Furthermore, a centralized system permits collaboration and operation execution by multiple financial institutions. Aside from these two scenarios, KYC processes can also be executed via a blockchain-based system. The blockchain's decentralized network would be highly transparent, facilitating the validation and verification of customer data in real-time for all relevant stakeholders. In addition, the immutability and cryptography of the blockchain ensure that client information is secure and immutable, thereby eradicating the risk of data breaches. Blockchain-based KYC can further improve the client experience by eliminating the requirement for redundant paperwork and document submissions. After banks grant consumers loans, a blockchain-based KYC system is proposed in this study to collect limit, risk, and collateral information from them. The approach built upon Ethereum grants financial institutions the ability to read and write financial data on the blockchain network. This KYC method establishes a transparent, dynamic, and expeditious framework among financial institutions. In addition, solutions are discussed for the Sybil attack, one of the most severe problems in such networks.Publication Restricted Utilizing Metaheuristics to Estimate Wind Energy Integration in Smart Grids With A Comparative Analysis of Ten Distributions(Taylor & Francis Inc., 2024) Wadi, Mohammed; ELMASRY, WİSAM; Çolak, İlhami; Jouda, Muhammed; Küçük, İsmailRenewable energy presents the most favorable approach to address the escalating challenge of greenhouse gas emissions while simultaneously guaranteeing the safeguarding of the environment. This article utilizes ten different distributions to approximate the wind energy integration in smart grids. The employed distributions are Rayleigh, Poisson, Weibull, Normal, Gamma, Laplace, LogNormal, Nakagami, Birnbaum Saunders, and Burr. The parameters of each distribution are calculated based on metaheuristic methods such as particle swarm optimization and genetic algorithms. Six error criteria have been employed to evaluate the precision of introduced distributions and metaheuristic methods. The approximation is performed by utilizing the wind data collected over three years hourly in the Marmara region of Turkiye. The empirical findings indicate that Gamma, Burr, and Weibull distributions exhibit more significant superiority than the remaining distributions across all datasets.Publication Open Access NeuroBioSense: A Multidimensional Dataset for Neuromarketing Analysis(Elsevier, 2024) KOCAÇINAR, BÜŞRA; İNAN, PELİN; ZAMUR, ELA NUR; ÇALŞİMŞEK, BUKET; AKBULUT, FATMA PATLAR; Çatal, ÇağatayIn the context of neuromarketing, sales, and branding, the investigation of consumer decision-making processes presents complex and intriguing challenges. Consideration of the effects of multicultural influences and societal conditions from a global perspective enriches this multifaceted field. The application of neuroscience tools and techniques to international marketing and consumer behavior is an emerging interdisciplinary field that seeks to understand the cognitive processes, reactions, and selection mechanisms of consumers within the context of branding and sales. The NeuroBioSense dataset was prepared to analyze and classify consumer responses. This dataset includes physiological signals, facial images of the participants while watching the advertisements, and demographic information. The primary objective of the data collection process is to record and analyze the responses of human subjects to these signals during a carefully designed experiment consisting of three distinct phases, each of which features a different form of branding advertisement. Physiological signals were collected with the Empatica e4 wearable sensor device, considering non-invasive body photoplethysmography (PPG), electrodermal activity (EDA), and body temperature sensors. A total of 58 participants, aged between 18 and 70, were divided into three different groups, and data were collected. Advertisements prepared in the categories of cosmetics for 18 participants, food for 20 participants, and cars for 20 participants were watched. On the emotion evaluation scale, 7 different emotion classes are given: Joy, Surprise, anger, disgust, sadness, fear, and neutral. This dataset will help researchers analyse responses, understand and develop emotion classification studies, the relationship between consumers and advertising, and neuromarketing methods.Publication Open Access Neurophysiological and Biosignal Data for Investigating Occupational Mental Fatigue: MEFAR Dataset(Elsevier, 2024) Derdiyok, Şeyma; AKBULUT, FATMA PATLAR; Çatal, ÇağatayThe prevalence of mental fatigue is a noteworthy phenomenon that can affect individuals across diverse professions and working routines. This paper provides a comprehensive dataset of physiological signals obtained from 23 participants during their professional work and questionnaires to analyze mental fatigue. The questionnaires included demographic information and Chalder Fatigue Scale scores indicating mental and physical fatigue. Both physiological signal measurements and the Chalder Fatigue Scale were performed in two sessions, morning and evening. The present dataset encompasses diverse physiological signals, including electroencephalogram (EEG), blood volume pulse (BVP), electrodermal activity (EDA), heart rate (HR), skin temperature (TEMP), and 3-axis accelerometer (ACC) data. The NeuroSky MindWave EEG device was used for brain signals, and the Empatica E4 smart wristband was used for other signals. Measurements were carried out on individuals from four different occupational groups, such as academicians, technicians, computer engineers, and kitchen workers. The provision of comprehensive metadata supplements the dataset, thereby promoting inquiries about the neurophysiological concomitants of mental fatigue, autonomic activity patterns, and the repercussions of a cognitive burden on human proficiency in actual workplace settings. The accessibility of the aforementioned dataset serves to facilitate progress in the field of mental fatigue research while also laying the groundwork for the creation of customized fatigue evaluation techniques and interventions in diverse professional domains.Publication Open Access Web Service Discovery: Rationale, Challenges, and Solution Directions(Elsevier, 2023) TOKMAK, AHMET VEDAT; AKBULUT, AKHAN; Çatal, ÇağatayService Oriented Architecture (SOA) is a methodology that promotes cooperation between services with diverse, but connected functions. Web Service technology paved the way for microservice architecture as it is a feature of modern web applications that resulted from the rise of SOA. With the proliferation of self-contained services, the ease of finding has emerged as a critical concern. Due to the increasing number of services that perform identical tasks, it has become difficult for users to select the most feasible service. Providing the most relevant service for the customer quickly is a crucial infrastructure task, and undiscovered services increase ecosystem expenses. Syntactic, semantic-conscious, and ontology-based studies have been presented as ways to improve the effectiveness and quality of service discovery techniques. While there are many approaches that have been proposed and validated for service discovery in literature, these studies are fragmented and there is a lack of overview of the techniques of web service discovery. As such, we conduct a Systematic Literature Review (SLR) study to review the existing body of knowledge surrounding service discovery and discuss the state-of-the-art. We present an overview of the techniques and empirical evidence by identifying, analyzing, and classifying the papers. Among the 764 papers we retrieved, 54 papers were included. We provide a comprehensive analysis of methodologies and tools for discovering web services.Publication Open Access Rüzgâr Enerjisi Potansiyelini Değerlendirirken Önemli Hususlar(Oğuzhan Yılmaz, 2023) Wadi, Mohammed; ELMASRY, WİSAM; Tamyiğit, Furkan AhmetRüzgâr rejimi dağılım modelinin belirlenmesi birkaç nedenden dolayı gereklidir, rüzgâr gücü çıktısını tahmin etmek en önemli konulardan biridir. Bu açıdan rüzgâr hızı dağılımını modellemek için Weibull, Gamma ve Rayleigh dağılımları en yaygın olarak kullanılan dağılımlardır. Ancak, tüm rüzgâr modellerini modellemede üstün olmayabilirler. Sonuç olarak, yerine geçecek dağılım fonksiyonlarının çalışılması gerekmektedir. Bu makale, rüzgâr hızı dağılımını tanımlamak için Weibull, Uç Değer, Ters Gauss, Lojistik, Log-Lojistik, Yarı-Normal, Burr Tipi XII, Genelleştirilmiş Uç Değer, Genelleştirilmiş Pareto ve T Konum-Ölçeği adlı on farklı dağılım fonksiyonlarını kapsamlı bir şekilde sunar. Ayrıca, her dağılımın parametre değerlerini optimize etmek için iki metasezgisel optimizasyon yöntemi olan Genetik Algoritması ve Parçacık Sürü Optimizasyonu kullanılmaktadır. Sunulan dağılımların iyi durumlarını (good-of-fitness) karşılaştırmak için yedi istatistiksel tanımlayıcı ile birlikte altı hata kriteri kullanılmıştır.Publication Metadata only Symptom Based Health Status Prediction via Decision Tree, KNN, XGBoost, LDA, SVM, and Random Forest(Springer Science and Business Media Deutschland GmbH, 2023) MERİÇ, ELİF; Özer, ÇaǧdaşMachine learning applications in health science become more important and necessary every day. With the help of these systems, the load of the medical staff will be lessened and faults because of a missing point, or tiredness will decrease. It should not be forgotten that the last decision lies with the professionals, and these systems will only help in decision-making. Predicting diseases with the help of machine learning algorithm can lessen the load of the medical staff. This paper proposes a machine learning model that analyzes healthcare data from a variety of diseases and shows the result from the best resulting algorithm in the model. It is aimed to have a system that facilitates the diagnosis of diseases caused by the density of data in the health field by using these algorithms of previously diagnosed symptoms, thus resulting in doctors going a faster way while diagnosing the disease and have a prediction about the diseases of people who do not have the condition to go to the hospital. In this way, it can ease the burden on health systems. The disease outcome corresponding to the 11 symptoms found in the data set used is previously experienced results. During the study, different ML algorithms such as Decision Tree, Random Forest, KNN, XGBoost, SVM, LDA were tried and compatibility/performance comparisons were made on the dataset used. The results are presented in a table. As a result of these comparisons and evaluations, it was seen that Random Forest Algorithm gave the best performance. While data was being processed, input parameters were provided to each model, and disease was taken as output. Within this limited resource, our model has reached an accuracy rate of 98%.Publication Metadata only Analysis of the Lingering Effects of COVID-19 on Distance Education(Springer Science and Business Media Deutschland GmbH, 2023) KOCAÇINAR, BÜŞRA; QARIZADA, NASIBULLAH; DİKKAYA, CİHAN; AZGUN, EMİRHAN; ELİF, YILDIRIM; AKBULUT, FATMA PATLAREducation has been severely impacted by the spread of the COVID-19 virus. In order to prevent the spread of the COVID-19 virus and maintain education in the current climate, governments have compelled the public to adopt online platforms. Consequently, this decision has affected numerous lives in various ways. To investigate the impact of COVID-19 on students’ education, we amassed a dataset consisting of 10,000 tweets. The motivations of the study are; (i) to analyze the positive, negative, and neutral effects of COVID-19 on education; (ii) to analyze the opinions of stakeholders in their tweets about the transition from formal education to e-learning; (iii) to analyze people’s feelings and reactions to these changes; and (iv) to analyze the effects of different training methods on different groups. We constructed emotion recognition models utilizing shallow and deep techniques, including Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long-short Term Memory (LSTM), Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM), and Logical Regression (LR). RF algorithms with a bag-of-words model outperformed with over 80% accuracy in recognizing emotions.Publication Metadata only Sentiment Analysis of Tweets on Online Education During COVID-19(Springer Science and Business Media Deutschland GmbH, 2023) YAZGAN, HARUN; ÖZBEK, ONUR; GÜNAY, AHMET CAN; AKBULUT, FATMA PATLAR; ELİF, YILDIRIM; KOCAÇINAR, BÜŞRA; ŞENGEL, ÖZNURThe global coronavirus disease (COVID-19) pandemic has devastated public health, education, and the economy worldwide. As of December 2022, more than 524 million individuals have been diagnosed with the new coronavirus, and nearly 6 million people have perished as a result of this deadly sickness, according to the World Health Organization. Universities, colleges, and schools are closed to prevent the coronavirus from spreading. Therefore, distance learning became a required method of advancing the educational system in contemporary society. Adjusting to the new educational system was challenging for both students and instructors, which resulted in a variety of complications. People began to spend more time at home; thus, social media usage rose globally throughout the epidemic. On social media channels such as Twitter, people discussed online schooling. Some individuals viewed online schooling as superior, while others viewed it as a failure. This study analyzes the attitudes of individuals toward distance education during the pandemic. Sentiment analysis was performed using natural language processing (NLP) and deep learning methods. Recurrent neural network (RNN) and one-dimensional convolutional neural network (1DCNN)-based network models were used during the experiments to classify neutral, positive, and negative contents.Publication Open Access Analysis of Facial Emotion Expression in Eating Occasions Using Deep Learning(Springer, 2023) ELİF, YILDIRIM; AKBULUT, FATMA PATLAR; Çatal, ÇağatayEating is experienced as an emotional social activity in any culture. There are factors that influence the emotions felt during food consumption. The emotion felt while eating has a significant impact on our lives and affects different health conditions such as obesity. In addition, investigating the emotion during food consumption is considered a multidisciplinary problem ranging from neuroscience to anatomy. In this study, we focus on evaluating the emotional experience of different participants during eating activities and aim to analyze them automatically using deep learning models. We propose a facial expression-based prediction model to eliminate user bias in questionnaire-based assessment systems and to minimize false entries to the system. We measured the neural, behavioral, and physical manifestations of emotions with a mobile app and recognize emotional experiences from facial expressions. In this research, we used three different situations to test whether there could be any factor other than the food that could affect a person’s mood. We asked users to watch videos, listen to music or do nothing while eating. This way we found out that not only food but also external factors play a role in emotional change. We employed three Convolutional Neural Network (CNN) architectures, fine-tuned VGG16, and Deepface to recognize emotional responses during eating. The experimental results demonstrated that the fine-tuned VGG16 provides remarkable results with an overall accuracy of 77.68% for recognizing the four emotions. This system is an alternative to today’s survey-based restaurant and food evaluation systems.Publication Open Access Investigating the Effects of Stress on Achievement: BIOSTRESS Dataset(Elsevier Inc., 2023) ÇÖPÜRKAYA, ÇAĞLA; MERİÇ, ELİF; ERİK, ELİF BERRA; KOCAÇINAR, BÜŞRA; AKBULUT, FATMA PATLAR; Çatal, ÇağatayThe effects of chronic stress on academic and professional achievement can have a substantial impact. This relationship is highlighted through a dataset that includes questionnaires and physiological data from a group of individuals. The questionnaire data of 48 individuals, the physiological data of 20 individuals during sessions with a psychologist, and the exam data of 8 individuals were analyzed. The questionnaire data collected includes demographic information and scores on the TOAD stress scale. Physiological data was captured using the Empatica e4, a wearable device, which measured various signals such as blood volume pulse, electrodermal activity, body temperature, interbeat intervals, heart rate, and 3-axis accelerometer data. These measurements were taken under different stress conditions, both high and low, during therapy sessions and an exam respectively. Overall, this study significantly contributes to our understanding of how stress affects achievement. By providing a large dataset consisting of questionnaires and physiological data, this research helps researchers gain a better understanding of the complex relationship between stress and achievement. It also enables them to develop innovative strategies for managing stress and enhancing academic and professional success.Publication Open Access A Comparative Assessment of Five Different Distributions Based on Five Different Optimization Methods for Modeling Wind Speed Distribution(Gazi University, 2023) Wadi, Mohammed; ELMASRY, WİSAMDetermining wind regime distribution patterns is essential for many reasons; modelling wind power potential is one of the most crucial. In that regard, Weibull, Gamma, and Rayleigh functions are the most widely used distributions for describing wind speed distribution. However, they could not be the best for describing all wind systems. Also, estimation methods play a significant role in deciding which distribution can achieve the best matching. Consequently, alternative distributions and estimation methods are required to be studied. An extensive analysis of five different distributions to describe the wind speeds distribution, namely Rayleigh, Weibull, Inverse Gaussian, Burr Type XII, and Generalized Pareto, are introduced in this study. Further, five metaheuristic optimization methods, Grasshopper Optimization Algorithm, Grey Wolf Optimization, Moth-Flame Optimization, Salp Swarm Algorithm, and Whale Optimization Algorithm, are employed to specify the optimum parameters per distribution. Five error criteria and seven statistical descriptors are utilized to compare the good-of-fitness of the introduced distributions. Therefore, this paper provides different important methods to estimate the wind potential at any site..Publication Open Access Design and Implementation of a Deep Learning-Empowered m-Health Application(Springer, 2023) AKBULUT, AKHAN; DESOUKI, SARA; ABDELKHALIQ, SARA; KHANTOMANI, LAYAL; Çatal, ÇağatayMany people are unaware of the severity of melanoma disease even though such a disease can be fatal if not treated early. This research aims to facilitate the diagnosis of melanoma disease in people using a mobile health application because some people do not prefer to visit a dermatologist due to several concerns such as feeling uncomfortable by exposing their bodies. As such, a skincare application was developed so that a user can easily analyze a mole at any part of the body and get the diagnosis results quickly. In the first phase, the corresponding image is extracted and sent to a web service. Later, the web service classifies using the pre-trained model built based on a deep learning algorithm. The final phase displays the confidence rates on the mobile application. The proposed model utilizes the Convolutional Neural Network and provides 84% accuracy and 72% precision. The results demonstrate that the proposed model and the corresponding mobile application provide remarkable results for addressing the specified health problem.Publication Restricted Multi-Modal Fusion Learning Through Biosignal, Audio, and Visual Content for Detection of Mental Stress(Springer London Ltd., 2023) GÜLİN, DOĞAN; AKBULUT, FATMA PATLARMental stress is a significant risk factor for several maladies and can negatively impact a person's quality of life, including their work and personal relationships. Traditional methods of detecting mental stress through interviews and questionnaires may not capture individuals' instantaneous emotional responses. In this study, the method of experience sampling was used to analyze the participants' immediate affective responses, which provides a more comprehensive and dynamic understanding of the participants' experiences. WorkStress3D dataset was compiled using information gathered from 20 participants for three distinct modalities. During an average of one week, 175 h of data containing physiological signals such as BVP, EDA, and body temperature, as well as facial expressions and auditory data, were collected from a single subject. We present a novel fusion model that uses double-early fusion approaches to combine data from multiple modalities. The model's F1 score of 0.94 with a loss of 0.18 is very encouraging, showing that it can accurately identify and classify varying degrees of stress. Furthermore, we investigate the utilization of transfer learning techniques to improve the efficacy of our stress detection system. Despite our efforts, we were unable to attain better results than the fusion model. Transfer learning resulted in an accuracy of 0.93 and a loss of 0.17, illustrating the difficulty of adapting pre-trained models to the task of stress analysis. The results we obtained emphasize the significance of multi-modal fusion in stress detection and the importance of selecting the most suitable model architecture for the given task. The proposed fusion model demonstrates its potential for achieving an accurate and robust classification of stress. This research contributes to the field of stress analysis and contributes to the development of effective models for stress detection.Publication Open Access Stacking-Based Ensemble Learning for Remaining Useful Life Estimation(Springer, 2023) Türe, Begüm Ay; AKBULUT, AKHAN; Zaim, Abdul Halim; Çatal, ÇağatayExcessive and untimely maintenance prompts economic losses and unnecessary workload. Therefore, predictive maintenance models are developed to estimate the right time for maintenance. In this study, predictive models that estimate the remaining useful life of turbofan engines have been developed using deep learning algorithms on NASA's turbofan engine degradation simulation dataset. Before equipment failure, the proposed model presents an estimated timeline for maintenance. The experimental studies demonstrated that the stacking ensemble learning and the convolutional neural network (CNN) methods are superior to the other investigated methods. While the convolution neural network (CNN) method was superior to the other investigated methods with an accuracy of 93.93%, the stacking ensemble learning method provided the best result with an accuracy of 95.72%.Publication Open Access Boosting the Visibility of Services in Microservice Architecture(Springer, 2023) TOKMAK, AHMET VEDAT; AKBULUT, AKHAN; Çatal, ÇağatayMonolithic software architectures are no longer sufficient for the highly complex software-intensive systems, which modern society depends on. Service Oriented Architecture (SOA) surpassed monolithic architecture due to its reusability, platform independency, ease of maintenance, and scalability. Recent SOA implementations made use of cloud-native architectural approaches such as microservice architecture, which has resulted in a new challenge: the discovery difficulties of services. One way to dynamically discover and route traffic to service instances is to use a service discovery tool to locate the Internet Protocol (IP) address and port number of a microservice. In the event that replicated microservice instances are found to provide the same function, it is crucial to select the right microservice that provides the best overall experience for the end-user. Parameters including success rate, efficiency, delay time, and response time play a vital role in establishing a microservice's Quality of Service (QoS). These assessments can be performed by means of a live health-check service, or, alternatively, by making a prediction of the current state of affairs with the application of machine learning-based approaches. In this research, we evaluate the performance of several classification algorithms for estimating the quality of microservices using the QWS dataset containing traffic data of 2505 microservices. Our research also analyzed the boosting algorithms, namely Gradient Boost, XGBoost, LightGBM, and CatBoost to improve the overall performance. We utilized parameter optimization techniques, namely Grid Search, Random Search, Bayes Search, Halvin Grid Search, and Halvin Random Search to fine-tune the hyperparameters of our classifier models. Experimental results demonstrated that the CatBoost algorithm achieved the highest level of accuracy (90.42%) in predicting microservice quality.Publication Metadata only Cyberbullying Detection Through Deep Learning: A Case Study of Turkish Celebrities on Twitter(IOS Press, 2023) Karadağ, Bulut; AKBULUT, AKHAN; Zaim, Abdul HalimOne of the ways that celebs maintain their fame in the modern era is by posting updates and photos to social media platforms like Twitter, Instagram, and Facebook. Comments left on their posts, however, expose them to cyberbullying. Cyberbullying, as a form of electronic device-based harassment, negatively impacts the lives of individuals. Thirty famous people from the fields of acting, art, music, politics, sports, and writing were chosen for this research. These notable figures include the top five Twitter followers of Turkey in each demographic. Between December 2019 and December 2020, comment responses for each celebrity were collated. Using the Deep Learning model, we were able to detect abuse content with an accuracy of 89%. Additionally, the percentage of celebrities exposed to cyberbullying by group was presented.