Sürdürülebilir Kalkınma Amaçları

• 7
• 2
• 2
• 1
• 1

• MakaleAçık Erişim2026/r/munzur/6091
  Sentiment Analysis of Monkeypox Tweets Using a Hybrid CNN-BiGRU Based Deep Learning Model

  Can, Umit

  Recently, epidemics have disrupted people's lives. Following the COVID-19 pandemic, cases of monkeypox, a rare disease, began to appear worldwide. These incidents raised societal tension and anxiety. Understanding public attitudes, sentiments, and perceptions regarding the epidemic is essential for developing and implementing effective policies and countermeasures. Online social media platforms are essential venues for expressing feelings and views. In particular, Twitter provides valuable information for measuring societal sentiment. Machine learning and deep learning methods have recently been widely used in social network analysis. Deep learning models, which deliver strong performance, particularly on complex and nonlinear problems, are used in natural language processing and text mining. In this study, a Convolutional Neural Network Bidirectional Gated Recurrent Unit (CNN-BiGRU) hybrid deep learning model was developed by combining a CNN and a BiGRU and applied for the monkeypox sentiment analysis. The dataset was labeled as positive, negative, or neutral with Valence Aware Dictionary and sEntiment Reasoner (VADER). In addition, FastText and Term Frequency-Inverse Document Frequency (TF-IDF) word-embedding methods were used for feature extraction, and the performance of machine-learning and deep-learning methods was evaluated separately. As a result of the experimental studies, the FastText-CNN-BiGRU model obtained the highest results with 0.9382 accuracy, 0.9347 macro precision, 0.9367 macro recall, 0.9355 macro F-score, and 0.9875 macro ROC-AUC. The FastText-CNN model was the second most successful, achieving 0.9303 accuracy, 0.9268 macro precision, 0.9280 macro recall, 0.9271 macro F-score, and 0.9862 macro ROC-AUC. The FastText-BiGRU model was the third most successful, achieving 0.9262 accuracy, 0.9224 macro precision, 0.9255 macro recall, 9236 macro F-score, and 0.9838 macro ROC-AUC.

  İncele
• MakaleAçık Erişim2026/r/munzur/5938
  Application of machine learning techniques for multi-class classification performance on mental disorders

  Can, Umit

  The contemporary social framework is predicated on a fiercely competitive system, which amplifies individual pressure and subsequently elevates the incidence of mental diseases, significantly affecting public health. Consequently, governmental and non-governmental groups allocate considerable resources to tackle these concerns. The precise identification of such illnesses via artificial intelligence techniques has become critically significant. These methodologies, widely used across several healthcare sectors, have been increasingly adopted in psychiatry. Evaluating psychiatric data to forecast disease accurately enables doctors to make better decisions and establishes a robust decision-support framework. This work employed prevalent machine learning techniques to identify mental illnesses utilizing a dataset with labels for seven distinct mental states—six disorders and one normative condition. The approaches were initially evaluated on an imbalanced dataset and subsequently on a balanced dataset, facilitating comparison of the outcomes. Among Categorical Boosting, Gradient Boosting, Extreme Gradient Boosting, Support Vector Machine, and Logistic Regression, the Support Vector Machine method demonstrated superior performance on both balanced and imbalanced datasets.

  • 
  İncele
• MakaleAçık Erişim2026Scopus SJRQ2/r/munzur/1302
  Translation-Aware Multilingual Sentiment Analysis: A Comparative Study of Machine Translation Systems and Transformer Ensembles

  Aslan, Serpil; Kartoğlu, Celal Şamil; Can, Ümit

  Multilingual sentiment analysis remains challenging because of cross-linguistic differences, uneven data availability, and potential shifts in meaning introduced during translation. Rather than proposing a new model, this study examines the performance of translation-based pipelines and ensemble Transformer models in multilingual settings. The primary objective is to investigate differences in downstream sentiment classification performance between translation systems within a consistent evaluation framework. Texts in Arabic, Chinese, French, and Italian are standardized through label harmonization and basic preprocessing and subsequently translated into English using Google Translate and the open-source LibreTranslate. Sentiment classification is performed using three Transformer-based models: DeBERTa-v3, RoBERTa, and BERTweet. The experiments are formulated as a binary polarity-classification task comprising positive and negative instances, with neutral samples excluded because their annotations are not directly comparable across the heterogeneous source datasets. Predictions from the individual models are combined through majority voting without introducing a new ensemble method. Experiments on the resulting binary datasets demonstrate that the ensemble achieves the highest mean weighted F1-score, with low fold-to-fold variation under both translation pipelines. Google Translate yields higher and more consistent classification performance across the evaluated subsets, whereas LibreTranslate exhibits greater variation. Given that the datasets differ in domain and genre, these variations cannot be attributed solely to linguistic differences. Overall, the findings reveal differences in downstream sentiment classification performance between the two translation pipelines and indicate that the ensemble combines high mean performance with low fold-to-fold variation within the evaluated binary setting.

  İncele
• MakaleAçık Erişim2026Scopus SJRQ1/r/munzur/1216
  Surface EMG-Based Hand Gesture Recognition Using a Hybrid Multistream Deep Learning Architecture

  Çelik, Yusuf; Can, Umit

  Surface electromyography (sEMG) enables non-invasive measurement of muscle activity for applications such as human–machine interaction, rehabilitation, and prosthesis control. However, high noise levels, inter-subject variability, and the complex nature of muscle activation hinder robust gesture classification. This study proposes a multistream hybrid deep-learning architecture for the FORS-EMG dataset to address these challenges. The model integrates Temporal Convolutional Networks (TCN), depthwise separable convolutions, bidirectional Long Short-Term Memory (LSTM)–Gated Recurrent Unit (GRU) layers, and a Transformer encoder to capture complementary temporal and spectral patterns, and an ArcFace-based classifier to enhance class separability. We evaluate the approach under three protocols: subject-wise, random split without augmentation, and random split with augmentation. In the augmented random-split setting, the model attains 96.4% accuracy, surpassing previously reported values. In the subject-wise setting, accuracy is 74%, revealing limited cross-user generalization. The results demonstrate the method’s high performance and highlight the impact of data-partition strategies for real-world sEMG-based gesture recognition.

  AtıfScopus4·WoS4·OpenAlex7

  İncele
• MakaleYalnız künye2026Scopus SJRQ1/r/munzur/1031
  Analysis of the impact of the social justice index on the COVID-19 pandemic using artificial intelligence in EU and OECD countries

  Utku, Anıl; Can, Ümit; BARUT, Abdulkadir; Pilatin, Abdulmuttalip; Sharabati, Abdel-Aziz Ahmad

  The COVID-19 disease has evolved into a pandemic that affects the world in multiple dimensions. While COVID-19 spread rapidly to several nations, distinct patterns of spread emerged in each nation. This spreading behavior is related to various characteristics of countries, including social, political, and economic factors. In this study, an analysis of the COVID-19 pandemic in EU and OECD countries was conducted, taking into account the social justice index for the first time. The impact and significance of the social justice index on the COVID-19 analysis results were attempted to be revealed. As a result, it is understood that the countries with high Social Justice Index scores and sub-dimension scores are more successful in combating the COVID-19 pandemic. In addition, a new LSTM-based model was developed to predict the number of cases and deaths in these countries. The designed LSTM approach was compared with some well-known deep learning and machine learning algorithms. As a result of the experimental studies, the LSTM model obtained the best prediction results. Finally, the similarities between these countries in terms of the characteristics of the COVID-19 pandemic, including death and case spread, were analyzed, and exciting results were disclosed.

  • 
  İncele
• MakaleYalnız künye2025/r/munzur/6173
  The Position of Metaheuristic Optimization Methods in Psychiatric Practice: A Bibliometric Analysis

  Tuğrul, Feride; Can, Umit

  Mental diseases affect billions of people worldwide to varying degrees. This poses a risk to public health. Although the levels and effects of diseases vary, in general, mental diseases prevent people from having a healthy relationship with themselves and their environment and prevent them from living their daily lives efficiently. Individuals who move away from the comfort of daily life face many difficulties in their social and professional lives. People with long-term mental disorders may also experience physiological disorders due to stress. When this situation is not treated and prevented, it becomes a public health problem. Therefore, centers such as health institutions, research centers, and universities operating in the field of health carry out many studies to detect and treat mental disorders. These methods include traditional methods, but today, they also include artificial intelligence-supported methods. AI-powered methods are used for early diagnosis and treatment of mental illnesses by analyzing large amounts of mental illness data of various types. Researchers are also using metaheuristic optimization algorithms to improve the performance of these methods. These optimization methods, which seek fast and high-performance solutions to very complex nonlinear problems, are increasing the success of machine learning and deep learning methods. Nonetheless, there is no distinct bibliometric analysis demonstrating the quantity and quality of research in this domain for scholars. This study aims to fill this gap in this field and to be a preliminary guide for researchers.

  AtıfTR Dizin1

  İncele
• MakaleYalnız künye2025Scopus SJRQ4/r/munzur/1948
  Deep learning-based hybrid traffic flow prediction model for freight transport

  Utku, Anıl; Kaya, Sema Kayapinar; Can, Umit

  Urban freight transportation is crucial to sustaining domestic and international trade and fulfilling the regular requirements of community companies and customers. Therefore, this study developed an efficient convolutional neural network (CNN) and long short-term memory (LSTM) based hybrid deep learning model for predicting traffic flow in freight transportation. The number of freight trucks that would travel along 663 Street in Anderlecht, Belgium, and their average speed was predicted. The created model was compared with support vector machine (SVM), linear regression (LR), multilayer perceptron (MLP), random forest (RF), LSTM, and CNN using the success measures. The critical finding of this study is that combining the feature extraction ability of the 1D CNN model with the learning ability of the LSTM model from multivariate traffic data increases the traffic density prediction success compared to other traditional methods. Based on the results of the experiments, the CNN-LSTM model outperformed all the other models employed in this study. With this model, freight density can be successfully predicted, assisting in planning freight transport activities.

  • 
  • 
  İncele
• MakaleAçık Erişim2025Scopus SJRQ2/r/munzur/915
  Advancing Air Quality Monitoring: Deep Learning-Based CNN-RNN Hybrid Model for PM2.5 Forecasting

  Utku, Anıl; Can, Umit; Alpsülün, Mustafa; Balıkçı, Hasan Celal; Amoozegar, Azadeh; Pilatin, Abdulmuttalip; Barut, Abdulkadir

  Particulate matter, particularly PM2.5, poses a significant threat to public health due to its ability to disperse widely and its detrimental impact on the respiratory and circulatory systems upon inhalation. Consequently, it is imperative to maintain regular monitoring and assessment of particulate matter levels to anticipate air pollution events and promptly mitigate their adverse effects. However, predicting air quality is inherently complex, given the multitude of variables that influence it. Deep learning models, renowned for their ability to capture nonlinear relationships, offer a promising approach to address this challenge, with hybrid architectures demonstrating enhanced performance. This study aims to develop and evaluate a hybrid model integrating Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for forecasting PM2.5 levels in India, Milan, and Frankfurt. A comparative analysis with established deep learning and machine learning techniques substantiates the superior predictive capabilities of the proposed CNN-RNN model. The findings underscore its potential as an effective tool for air quality prediction, with implications for informed decision-making and proactive intervention strategies to safeguard public health.

  • 
  AtıfScopus10·WoS6·OpenAlex11

  İncele
• MakaleAçık Erişim2024/r/munzur/6266
  Sentetik Atık Takviyeli Poliolefinlerin Mekanik Özelliklerinin GA-LSTM Hibrit Modeli ile Tahmini

  Utku, Anıl; Kismet, Yilmaz; Can, Umit

  Bu çalışmada, parçacık takviyeli termoplastiklerin enjeksiyon kalıplamasında kullanılan üretim parametrelerinin ürün kalitesi ve üretilen parçanın mekanik özellikleri üzerindeki etkileri optimize edilmiş bir Genetik Algoritma-Uzun Kısa Süreli Bellek (GA-LSTM) hibrit derin öğrenme yöntemi kullanılarak modellenmiştir. Burada termoplastik olarak poliolefinler grubunun en önemli üyesi olan AYPE, YYPE ve PP kullanılırken takviye elemanı olarak ise toz halde sentetik boya atıkları kullanılmıştır. Farklı parametreler kullanılarak enjeksiyon kalıplama yoluyla 819 numune üretilmiş ve her numune üzerinde mekanik çekme, üç nokta eğme ve izod darbe testleri gerçekleştirilmiştir. GA-LSTM modeli, kullanılan parametreler ve deneysel süreç boyunca elde edilen sonuçlarla eğitilmiş ve tahmin edilen değerlerin gerçek değerlere karşılık geldiği belirlenmiştir. Hibrit GA-LSTM modelinin başarısını ölçmek için iyi bilinen yöntemler kullanılmıştır. Elde edilen sonuçlara göre tasarlanan GA-LSTM modeli en iyi sonuçları üretmiştir.

  • 
  • 
  İncele
• DüzeltmeAçık Erişim2024Scopus SJRQ2/r/munzur/2084
  Correction to: Detection of hateful twitter users with graph convolutional network model (Earth Science Informatics, (2023), 16, 1, (329-343), 10.1007/s12145-023-00940-w)

  Utku, Anıl; Can, Umit; Aslan, Serpil

  In this article the affiliation details for Serpil Aslan were incorrectly given as ‘Software Engineering Department, Turgut Üzal University, 44210 Malatya, Turkey’ but should have been ‘Department of Software Engineering, Faculty of Engineering and Natural Sciences, Malatya Turgut Ozal University, Malatya, Turkey.’ The original article has been corrected.

  AtıfScopus1·OpenAlex2

  İncele