Yazar "Savaş, Serkan" seçeneğine göre listele

Listeleniyor 1 - 6 / 6
  • Küçük Resim
    Öğe
    Analysis of Digital Participation with Text Mining Method and 2017 Turkey Social Media Referendum
    (Istanbul Univ, 2022) Savaş, Serkan
    Social networks have recently become a guiding factor for communities in many areas. Twitter is one of the most important platforms on the social network for setting instant agendas among these networks. Especially in the last decade, social networks have been used in political processes globally. Social networks have become important platforms for political parties to manage their propaganda and for the public to express their opinion. Based on this, a text mining study was carried out on Twitter during Turkey's 2017 referendum process. In the study, public opinions posted on Twitter were examined to reveal the effectiveness of political processes. For this purpose, the ratios of yes/no opinions, political parties, and political party leaders were analyzed, which are the preferred options of the referendum. With these analyses, an empirical study on social-cyber intelligence was carried out, and the potential of decision support systems for policymakers was revealed
  • Küçük Resim
    Öğe
    Application of deep ensemble learning for palm disease detection in smart agriculture
    (Cell Press, 2024) Savaş, Serkan
    Agriculture has notably become one of the fields experiencing intensive digital transformation. Leveraging state-of-the-art techniques in this domain has provided numerous advantages for agricultural activities. Deep learning (DL) algorithms have proven beneficial in addressing various agricultural challenges. This study presents a comprehensive investigation into applying DL models for palm disease detection and classification in the context of smart agriculture. The research aims to address the limitations observed in previous studies and improve the robustness and generalizability of the results. To achieve this, a two-stage optimization methodology is employed. First, transfer learning and fine-tuning techniques are applied using various pretrained deep neural network models. The experiments show promising results, with all models achieving high accuracy rates during training and validation. Furthermore, their performance on unseen test data is also assessed to ensure practical applicability. The top-performing models are MobileNetV2 (92.48 %), ResNet (92.42 %), ResNetRS50 (92.30 %), and DenseNet121 (92.01 %). Second, a deep ensemble learning approach is applied to enhance the models' generalization capability further. The best-performing models with different criteria are combined using the ensemble technique, resulting in remarkable improvements in disease detection tasks. DELM1 emerges as the most successful ensemble model, achieving an ROC AUC Score of 99%. This study demonstrates the effectiveness of deep ensemble learning models in palm disease detection and classification for smart agriculture applications. The findings contribute to advancing disease detection systems and emphasize the potential of ensemble learning. The study provides valuable insights for future research, guiding the application of DL techniques to address critical agricultural challenges and improve crop health monitoring systems. Another contribution is combining various plant diseases and insect pest classes using diverse datasets. A comprehensive classification system is achieved by considering different disease classes and stages within the white scale category, improving the model's robustness.
  • [ X ]
    Öğe
    Skin lesion classification by weighted ensemble deep learning
    (Springer International Publishing, 2024) Al-Saedi, Doaa Khalid Abdulridha; Savaş, Serkan
    Skin cancer represents a significant global health threat with potentially fatal consequences if left undiagnosed. Early detection is crucial for successful patient treatment, yet accurate identification of skin lesions poses a challenge even for experienced dermatologists. In this context, the development of computer-aided skin lesion classification systems emerges as a promising path to empower dermatologists with the potential for earlier diagnoses and more effective treatment interventions. This study proposes a two-stage approach for early detection of skin cancer. Firstly, eight pre-trained deep architectures were tested on the ISIC dataset using transfer learning and fine-tuning. Secondly, three successful models with the highest accuracy were chosen, and ensemble learning was employed to obtain a final result. The ensemble learning method outperformed individual models, achieving a remarkable ROC AUC rate of 99.96%. DenseNet121 exhibited the highest performance among the individual models, with accuracy rates of 99.75%, 98.2%, and 99.6% for the train, validation, and test datasets, respectively. The promising results hold significant potential for early detection and treatment of skin cancer, a prevalent global disease. These findings could prove invaluable for clinics, offering valuable support to their decision-making processes and enhancing their ability to combat this widespread health concern. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.
  • [ X ]
    Öğe
    Smart Maintenance with Regression Analysis for Efficiency Improvement in Photovoltaic Energy Systems
    (University of Tehran, 2023) Ay, İlker; Kademli, Murat; Savaş, Serkan; Karellas, Sotirios; Markopoulos, Angelos; Hatzilau, Christina-Stavroula; Devlin, Philip
    This research had the overarching goal of optimizing maintenance intervals and reducing the maintenance workload by enhancing accessibility for individuals lacking technical expertise in the upkeep of photovoltaic systems, with a particular focus on rooftop applications. The study achieved this objective by employing a linear regression algorithm to analyse climatic parameters such as wind speed, humidity, ambient temperature, and light intensity, collected from the installation site of a photovoltaic solar energy system. Simultaneously, the current and voltage values obtained from the system were also examined. This analysis not only facilitated the determination of power generation within the system but also enabled real-time detection of potential issues such as pollution, shadowing, bypass, and panel faults on the solar panels. Additionally, an artificial intelligence-supported interface was developed within the study, attributing any decline in power generation to specific causes and facilitating prompt intervention to rectify malfunctions, thereby ensuring more efficient system operation. © The Author(s).
  • Küçük Resim
    Öğe
    The use of artificial intelligence in learning on social networking sites among Hungarian and Turkish youth
    (IEEE, 2024) Csiszarik-Kocsir, Agnes; Revak, Bemadett; Savaş, Serkan
    Digitalisation, and the modernisation that comes with digitalisation, has become part of our lives. Alongside digitalisation, innovation has become the leading concept of the 21st century, and we are now seeing the emergence of artificial intelligence. The rise of artificial intelligence has become an unstoppable process. We see it in the day-to-day management of our business, in customer services, but we are now seeing artificial intelligence applications replacing live humans on the screen. Encountering artificial intelligence is a phenomenon that awaits us regardless of our age. The aim of our research is to investigate how and in what form young people, who are part of the education system, encounter AI. We wanted to assess, based on the results of a questionnaire survey conducted in Hungary and Turkey, where and in what form young people in the age group 14 and above encounter any form of AI. We used cross tabulation analysis to draw conclusions in order to get a useful picture of the place and role of AI in the lives of young people.
  • [ X ]
    Öğe
    Tüm Yönleriyle Metaverse Çalışmaları, Teknolojileri ve Geleceği
    (2022) Güler, Osman; Savaş, Serkan
    İnsan hayatındaki beşinci boyutu olarak adlandırılan siber uzay, metaverse uygulamaları ile birlikte daha farklı bir yaşam alanı haline dönüşmeye başlamıştır. Mevcut haliyle eğitim, oyun ve eğlence, sanat, mimari, ticari vs. uygulamaların, alan ihtiyaçlarına yönelik kullanılmakta olduğu metaverse evreni, son dönemde teknoloji lideri kuruluşların odaklandığı ve daha fazla yatırım yaptığı bir alan haline gelmiştir. Bu yatırımlarla birlikte metaverse evrenine yönelik 3 boyutlu teknolojiler, sanal gerçeklik, artırılmış gerçeklik ve yapay zekâ çalışmaları da artış göstermiştir. Bu konuda yaşanılan gelişmelerden yola çıkarak bu çalışma, metaverse evrenindeki çalışmaların önemi ve gerekliliğini vurgulamaktadır. Çalışmada metaverse ile ilgili araştırmalar ve sonuçları incelenmiştir. Bu amaçla betimsel bir araştırma modeli kullanılmıştır. Araştırma felsefesi açısından uygulama yöntemine ilişkin temel araştırma modeli ve belge araştırma modeli oluşturulmuştur. Araştırmanın evrenini Google Akademik üzerinden elde edilen çalışmalar oluşturmaktadır. Belirlenen anahtar kelime ile metaverse konusunda son 20 yılda yapılan çalışmalar indirilmiştir. Araştırma sonucunda metaverse ile ilgili çalışmaların özellikle pandemi sonrasında artış gösterdiği, özellikle oyun sektöründe kullanıldığı ve giderek artan farklılıkta alanlara da uygulandığı tespit edilmiştir. Milyarlarca kullanıcıya ulaşması beklenen bu evrende yer almak isteyen ülkelerin çalışmalarına hız kazandırması ve “kullanıcı” ile “üretici” pozisyonlarından hangisinde yer almak istediğine karar vermesi gerektiği belirtilmiştir.