International Journal of Advanced Natural Sciences and Engineering Researches

Mediation and Conflict Prevention in Hungarian Public Service: An Analysis of Training Guidelines and Policy Frameworks

Hossain MD Shakhawoat — Fri, 05 Dec 2025 00:00:00 +0100

Mediation and conflict prevention have become increasingly important components of contemporary public administration, particularly in systems undergoing modernization and public sector reform. In Hungary, various policies, competency frameworks, and training programs have been introduced to strengthen the professional capacities of public servants; however, the actual integration of mediation principles and conflict-prevention skills within these institutional documents has not been systematically analyzed. This study conducts a qualitative content analysis of key national public service training guidelines, public administration development strategies, sectoral policy frameworks, and ethical codes to explore how these documents articulate the concepts, expectations, and practical approaches related to conflict prevention. The analysis identifies several emerging trends, including a growing emphasis on communication skills, collaborative problem-solving, ethical conduct, and leadership responsibility in managing organizational and community-level conflicts. At the same time, the findings reveal significant inconsistencies across institutions in how mediation is defined, operationalized, and embedded within training structures. Many documents emphasize general behavioral competencies but lack concrete methodological guidance, standardized mediation protocols, or alignment with international and EU mediation standards. The study argues that despite positive elements, Hungarian public service training and policy frameworks currently present a fragmented picture of mediation-related competencies. To address these gaps, the paper recommends developing a unified national mediation competency framework, integrating practical and scenario-based training elements, and harmonizing domestic standards with international best practices. Strengthening these areas would enhance the conflict-prevention capacity of public servants and contribute to a more responsive, professional, and citizen-focused public administration.

E-Governance as a Tool for Transparent and Peaceful Public Administration: Evidence from Bangladesh

Fatema Akter — Fri, 05 Dec 2025 00:00:00 +0100

This study explores the role of e-governance as both a communication and service-delivery tool that enhances transparency, citizen trust, efficiency, and peaceful public administration in Bangladesh. The research adopts a mixed-methods approach, integrating quantitative survey data (n = 300) with a municipal-level case study (n = 120 service cases) to provide a comprehensive analysis of how digital governance initiatives influence citizen perceptions and administrative performance. The survey investigates public attitudes toward e-governance, focusing on perceived transparency, trust, accessibility, and responsiveness, while the case study examines measurable improvements in service delivery, including processing times, error rates, and citizen satisfaction. Findings indicate moderate levels of citizen trust in e-governance platforms, with a mean trust score of 3.26 out of 5. Analysis of service efficiency shows significant reductions in average service time, from 12.45 days prior to implementation to 8.38 days post-adoption. Additionally, the study identifies a positive relationship between education level and perceived transparency and efficiency, highlighting the importance of digital literacy in maximizing the benefits of e-governance. Qualitative insights from the case study reveal that clear communication, timely updates, and interactive digital tools contribute substantially to citizen satisfaction and confidence in public administration. The research further considers the potential of emerging innovations, including artificial intelligence, to enhance the predictability, personalization, and transparency of public services. AI-driven tools, such as automated information portals, chatbots, and predictive analytics, are discussed as mechanisms for improving both efficiency and citizen engagement. The study concludes with policy recommendations for integrating digital innovations into Bangladesh’s public administration, emphasizing citizen-centered design, transparency, and participatory communication strategies. These findings contribute to the broader discourse on digital governance and provide actionable insights for policymakers aiming to strengthen trust, efficiency, and stability in public service delivery.

Caputo-Fabrizio Fractional Elzaki Decomposition Technique for Caputo Fabrizio Fractional Telegraph Equation

Aslı ALKAN, Salih TÜRK, Halil ANAÇ — Wed, 03 Dec 2025 00:00:00 +0100

This study introduces the Caputo–Fabrizio fractional Elzaki decomposition technique (CFFEDT)
for addressing time-fractional telegraph equation. The approach integrates the Caputo–Fabrizio operator
with the Elzaki transform to obtain efficient and precise solutions. Quantitative illustrations validate the
efficacy of the suggested method.

Cadmium and Chrome concentration levels in Red mullet and European hake from Albanian coastal waters

Wed, 03 Dec 2025 00:00:00 +0100

This study investigates the accumulation levels of cadmium (Cd) and chromium (Cr) in edible
tissue of two commercially important fish species, Red mullet and European hake sampled from the
Albanian coastal waters. Fish samples were classified in two size groups, small and large fish size.
Concentration levels of Cd and Cr expressed in mg kg wet weight were measured by using absorption
spectrophotometry. The average mean concentration levels of Cd and Cr were determined and compared
between small and large individuals to assess possible size-dependent bioaccumulation patterns. Results
indicate slightly higher mean Cd and Cr concentration levels in larger fish size of both fish species, though
differences were not statistically significant (p > 0.05). The measured values of cadmium resulted in both
fish size higher than the maximum permissible limits established by the European Commission for human
consumption, suggesting health risk to local consumers. Continued monitoring of both fish species is
recommended to assess potential changes in metal bioaccumulation linked to regional contamination
sources.

Numerical Comparison of Thermal Performances of Different Insulation Materials in External Wall Model

Mehmet KAN — Wed, 03 Dec 2025 00:00:00 +0100

In terms of heating and cooling buildings, thermal insulation is one of the most effective ways
to conserve energy. The selection of the insulation material and the determination of the optimal insulation
thickness are very important in terms of energy savings and providing thermal comfort conditions. In this
study, the thermal behavior of wall models designed using 4 different insulation materials for an exterior
wall was examined. The analysis was carried out using Ansys software for 4 different wall models with
brick as building material and Expanded Polystyrene (EPS), Extruded Polystyrene (XPS), Neopor and
Polyisocyanurate (PIR) as insulation materials. Wall models consist of 20 mm internal plaster, 215 mm
brick, insulation material (EPS, XPS, Neopor and PIR) and 30 mm external plaster. To compare the analysis
results, the insulation materials thickness was determined to be 100 mm. The temperature contours obtained
for these wall models were compared to the optimal insulation thicknesses and payback periods obtained
from the calculations. Additionally, costs were examined from an economic perspective. It was observed
that heat loss was minimal, and temperature curves were better in the wall model created using PIR.
Furthermore, an examination of optimal insulation thicknesses and payback periods revealed that PIR
insulation thickness was more optimal, but despite its higher unit cost, the payback period was shorter.

Investigation of the Thermal Performance of Heat Sinks with Different Fins

Mehmet KAN — Wed, 03 Dec 2025 00:00:00 +0100

Operating temperature increases reduce processor efficiency, which can have a detrimental
effect on system performance. The thermal performance of heat sinks with various designs for processor
cooling were compared in this study. Using phase change materials (PCM) for cooling is a great way to
keep electronic devices from overheating. However, PCM’s low thermal conductivity limits a heat sink's
ability to transfer heat. Phase change material (PCM) was inserted into three distinct heat sink
configurations long fin (LF), short fin (SF), and original fin (OF) and analyses were conducted to determine
how these configurations affected thermal performance. In the current study, PCM (paraffine) was used to
investigate the thermal performance results of the three heat sinks at a heating power of 30 W. This allowed
for the transfer of heat between the designed heat sinks. A fluid volume model was used to track the working
fluid inside the heat sink. The phenomenon of heat transfer that occurs when solid paraffine melts was
examined. Because of the largest heat transfer area between the PCM, the results indicated that the PCM
heat sink with OF exhibited optimal thermal performance. Furthermore, the full melt time of the PCM heat
sink with OF is 20 % shorter than that with other fin shapes.

Sağlık Personellerine Yönelik KBRN Eğitimlerinin Etkililiği: Farkındalık, Beceri ve Hazırlık Düzeyleri Üzerine Güncel Bir Derleme

Abdullah ÖZKAN, Mustafa Okan AKATAY — Wed, 03 Dec 2025 00:00:00 +0100

Sağlık personellerinin Kimyasal, Biyolojik, Radyolojik ve Nükleer (KBRN) olaylara etkin biçimde
müdahale edebilmesi, yalnızca bireysel güvenliğin korunması açısından değil, aynı zamanda sağlık
sistemlerinin olağan işleyişinin olay anında ciddi bir kesintiye uğramamasını sağlamak bakımından da
önem taşımaktadır. Son yıllarda yapılan araştırmalar, birçok kurumda sağlık çalışanlarının KBRN
konusundaki başlangıç düzeyindeki farkındalıklarının ve özellikle teknik uygulamalara ilişkin becerilerinin
sınırlı olduğunu göstermektedir. Bu eksiklik, olayların beklenmedik biçimde geliştiği ve karar verme
süresinin kısıtlı olduğu durumlarda daha görünür hâle gelmektedir. Özellikle kişisel koruyucu donanımın
doğru kullanılmaması, dekontaminasyon adımlarının atlanması veya yanlış uygulanması gibi hatalar, bu
yetersizliğin sahadaki en somut yansımaları arasında sayılmaktadır. Bununla birlikte, yapılandırılmış ve
düzenli şekilde uygulanan eğitim programlarının hem teorik bilgi düzeyinde hem pratik becerilerde hem de
genel hazırlık düzeyinde dikkate değer iyileşmeler oluşturduğu çeşitli çalışmalarda ifade edilmektedir.
Bu derleme, 2010 sonrasında yayımlanan ulusal ve uluslararası literatürü inceleyerek KBRN eğitimlerinin
sağlık personeli üzerindeki etkilerini karşılaştırmalı bir yaklaşımla değerlendirmektedir. Bulgular,
simülasyon temelli eğitimlerin, çok paydaşlı tatbikatların ve karma öğrenme modellerinin özellikle beceri
transferi, karar verme doğruluğu ve güvenli uygulama açısından öne çıktığını ortaya koymaktadır. Öte
yandan, bazı araştırmalarda eğitimle kazanılan bilgi ve becerilerin zaman içinde azaldığı da
belirtilmektedir. Bu durum, periyodik güncelleme ihtiyacını ve kurumların ölçme-değerlendirme
süreçlerini güçlendirme gereğini daha görünür kılmaktadır. Ayrıca kurum içi koordinasyon, zaman kısıtları
ve personel devri gibi etkenlerin de eğitimlerin sürdürülebilirliğini zorlaştırdığı görülmektedir.

Professional Challenges Of Advanced Practice Anesthesia Nurses In Albania

Rezarta Tota, Xhejni Kurti, Arben Ndreu — Wed, 03 Dec 2025 00:00:00 +0100

Anesthesia nursing is a highly specialized field requiring advanced clinical knowledge and
technical skills. In Albania, formal anesthesia training is limited to a single undergraduate course, resulting
in limited preparedness and increased reliance on untrained staff. This study distributed a structured online
questionnaire to 24 anesthesiologists from five major hospitals, investigating competencies, perceived
roles, and institutional support for anesthesia nurses. Results highlight a lack of standardized protocols,
professional requirements, and structured educational pathways. A comparative analysis with international
standards, such as Austria’s one-year diploma program, reveals significant gaps. Urgent reforms are needed
in education, professional specialization, and protocol standardization to improve patient safety and
workforce competence.

Deep Learning Image Analysis for Pedestrian and Animal Safety in Traffic intersection using YOLOv10 and YOLOv11

Bayan Sheikh Omar, Önder Yakut — Wed, 03 Dec 2025 00:00:00 +0100

Ensuring the safety of pedestrians and animals at traffic intersections remains a critical
challenge for intelligent transportation systems, particularly in complex urban environments. Traditional
object detection methods often fail to deliver reliable real-time performance. However, recent
advancements in deep learning, particularly within the YOLO (You Only Look Once) family, have
significantly improved detection speed and accuracy. While earlier versions, such as YOLOv5 to
YOLOv9, have been extensively studied, the potential of the more recent YOLOv10 and YOLOv11
architectures remains largely unexplored. This study provides a comparative analysis of these models in
the context of detecting pedestrians and animals at intersections. The results show that YOLOv11
achieves an impressive mean Average Precision (mAP) of 89%, outperforming previous YOLO versions
and current state-of-the-art methods. This positions YOLOv11 as a strong candidate for real-time traffic
safety applications. The findings underscore the potential of cutting-edge deep learning models in
advancing traffic safety and contribute to the development of smarter, more responsive transport
infrastructure.

Global Trends in Ultrasonography Research in Veterinary Anatomy

Osman Yılmaz — Wed, 03 Dec 2025 00:00:00 +0100

This study presents a bibliometric analysis of global trends in ultrasonography (USG) studies
in veterinary anatomy. The aim of this study is to investigate publications related to USG studies in
veterinary anatomy worldwide and to analyse global trends and groups in this field. This bibliometric
study investigates ultrasound studies in veterinary anatomy conducted worldwide between 1983 and
2025. To this end, 715 studies were examined as a result of searches conducted in the Web of Science
(WOS) database using the keywords ‘veterinary anatomy, ultrasonography, USG’. After excluding
ineligible studies and non-article studies, 693 articles were analysed. Information such as title, author
names, publication year, journal name, and number of citations was used for data collection. All text data
was analysed using VOSviewer software to ensure accuracy and reliability. In this study, analyses using
text mining and data visualisation methods (e.g., bubble maps) helped to make the results more
understandable. A total of 13,452 citations were found for the 693 articles examined in the WOS
database. The average number of citations per article is 19, and the H-index is 53. Since 1993, there has
been a significant increase in both the number of articles and citations. The majority of the articles (87%)
were published in the fields of veterinary sciences, radiology and nuclear medicine, agriculture and
zoology, indicating the increasing prevalence of USG studies in these areas. The countries with the
highest number of publications are the USA, England, France and Germany (66.8%), with Turkey
ranking 16th in this area. The main keywords are "ultrasonography, ultrasound, sonography, anatomy,
computed tomography, imaging", indicating that the studies focus particularly on these topics. Our study
shows that there are many researchers active in the field of veterinary anatomy and that the number of
studies in this area is increasing. This bibliometric analysis reveals global trends and significant studies in
the field of veterinary anatomy and ultrasound imaging, providing valuable insights into the future
direction of research in this area.

INSEAN E779A Pervanesi yakın alan su altı gürültüsü yayılımının sayısal olarak incelenmesi

Giray Esmer, Taner Coşgun — Wed, 03 Dec 2025 00:00:00 +0100

Bu çalışmada, bir referans gemi pervanesi olan INSEAN E779A geometrisi etrafındaki akış alanı ve
gürültü yayılımı sayısal olarak incelenmiştir. Bu kapsamda, hesaplamalı akışkanlar dinamiği yardımıyla
gerçekleştirilen sayısal çözümlerde Reynolds Ortalamalı Navier–Stokes (RANS) tabanlı akış çözümü ile
Ffowcs Williams–Hawkings (FWH) akustik anoloji yöntemi birlikte kullanılmıştır. Artan deniz trafiğiyle
birlikte artış gösteren gemi kaynaklı su altı gürültü kirliliği deniz ekosistemine olumsuz etkileri nedeniyle
dikkatle değerlendirilmektedir. Konu üzerine uygulamaya alınan uluslararası düzenlemelerin de etkisiyle
gemilerden yayılan su altı gürültüsü önemi gittikçe artan bir çalışma alanı haline gelmektedir. Bu
çalışmanın amacı, uygulanan sayısal yöntemin sonuçların doğruluğu üzerine etkinliğinin test edilmesidir.
Sabit pervane ilerleme katsayısında gerçekleştirilen incelemede farklı konumlardaki alıcı noktalarındaki
sayısal sonuçlar literatürle karşılaştırmalı olarak sunulmuştur. Elde edilen sonuçlar, RANS ile birlikte
kullanılan FWH anolojisinin yakın alan pervane gürültüsünü başarılı biçimde modelleyebildiğini ortaya
koymaktadır. Yakın alan basınç davranışının güvenilir şekilde modellenmesinin uzak alan URN tahminleri
için kritik bir temel oluşturmaktadır.

Enhancing Thematic Classification and Semantic Consistency in Albanian Text Using Synthetic Data and Transformer-Based Embedding Models

Leona Rexhaj, Luela Prifti — Wed, 03 Dec 2025 00:00:00 +0100

This paper proposes a complete computational framework for the thematic classification and semantic
analysis of Albanian text, focusing on Mathematical Sciences. Due to the scarcity of structured and annotated
Albanian academic corpora, the study introduces a data augmentation approach combining real student theses
with synthetic texts generated through large language models (LLMs). The thematic distribution extracted from
the real dataset shows that the dominant academic sections are Rezultate (463), Perfundime (357), Hyrje (117)
and Metodologji (63). After integrating synthetic material, most sections increase substantially, particularly
methodological and theoretical components, while Rezultate decreases slightly due to the more theoretical
nature of synthetic documents. Semantic coherence was evaluated using cosine similarity with the multilingual
E5-Large encoder. For the real dataset, the strongest semantic alignment was found between Rezultate -
Përfundime (0.8171), followed by Hyrje -Metodologji (0.7923) and Analize teorike - Modelim matematik
(0.7864), confirming the expected structural and conceptual proximity between these academic sections. Post
augmentation values remained comparable at 0.8101, 0.7881, and 0.7902 respectively. These results
demonstrate that the generated synthetic material preserves academic structure and maintains coherent thematic
relationships. Transformer-generated synthetic data can enrich Albanian academic corpora without degrading
semantic or structural consistency. The proposed framework provides a replicable approach for large-scale text
processing in low-resource languages, supporting future research tasks such as document categorization,
academic writing analysis, and digital humanities.

Early Detection of Heart Diseases through Deep Learning–Based Electrocardiogram (ECG) Analysis

İsa Ataş, Bashar Alhajahmad, Musa Ataş — Wed, 03 Dec 2025 00:00:00 +0100

Cardiovascular diseases are among the leading causes of death worldwide, and early diagnosis
plays a critical role in reducing mortality rates. Electrocardiogram (ECG) signals are a fundamental
diagnostic tool for identifying arrhythmias; however, manual interpretation requires expert knowledge and
is prone to diagnostic errors. In this study, a hybrid deep learning model combining Convolutional Neural
Networks (CNN) and Long Short-Term Memory (LSTM) networks was developed for automatic
arrhythmia detection from ECG signals. The model was trained using the MIT-BIH Arrhythmia Database,
and several preprocessing steps—including noise removal, windowing, standardization, and SMOTE-based
data balancing—were applied. The proposed CNN-LSTM model achieved high performance in five-class
(N, S, V, F, Q) arrhythmia classification, with an accuracy of 95.2%, an F1 score of 92%, and a loss value
of 0.14. The confusion matrix results indicate that the model demonstrates strong generalization capability,
particularly for the N, V, and Q classes. These findings suggest that the hybrid architecture can serve as an
effective decision-support tool in ECG signal analysis. Furthermore, integrating the proposed system into
portable devices may enable real-time and automated monitoring of cardiac rhythms; however, the model
has not been approved by regulatory agencies such as the FDA or CE, and further validation is required
before clinical use. Additionally, by reducing the diagnostic burden on clinicians, the system has the
potential to improve both accuracy and efficiency within clinical workflows. Future studies incorporating
cross-dataset validation and larger ECG collections may further strengthen the reliability and robustness of
the proposed approach.

Bilgisayarlı Görüntü İşleme: Görsel Bilginin Modellemesi ve Çözümlenmesi

İsa Ataş, Musa Ataş, Bashar Alhajahmad — Wed, 03 Dec 2025 00:00:00 +0100

Bilgisayarlı görüntü işleme, sayısal olarak temsil edilen görsel verilerin modellenmesi, işlenmesi
ve semantik açıdan anlamlandırılmasını amaçlayan matematiksel, istatistiksel ve yapay öğrenme tabanlı
yöntemleri kapsayan geniş ve disiplinlerarası bir araştırma alanıdır. Günümüzde bu alan, özellikle derin
öğrenme tekniklerinin katkısıyla hem düşük seviyeli görüntü işleme görevlerinde (gürültü giderme, kenar
tespiti, filtreleme ve iyileştirme) hem de yüksek seviyeli kavramsal çıkarımlarda (nesne tanıma, sahne
analizi, anomali tespiti ve görüntü sınıflandırma) kayda değer ilerlemeler kaydetmiştir. Bu çalışmada,
dijital görüntü modelleme yaklaşımları, özellik çıkarımı yöntemleri, segmentasyon teknikleri ve nesne
tanıma süreçleri ayrıntılı bir biçimde incelenmiş, özellikle konvolüsyonel sinir ağları, dikkat
mekanizmaları ve transfer öğrenmeye dayalı derin öğrenme yöntemlerinin görüntü analizi üzerindeki
etkileri vurgulanmıştır. Ayrıca, bilgisayarlı görüntü işlemenin sağlık görüntüleme, otonom sistemler,
güvenlik ve gözetim, endüstriyel kalite kontrol gibi çeşitli uygulama alanlarındaki önemi ve potansiyeli
ele alınmıştır. Çalışmada mevcut yöntemlerin karşılaştığı başlıca zorluklar, veri çeşitliliği, büyük veri
gereksinimi, model açıklanabilirliği, hesaplama maliyetleri ve gerçek zamanlı uygulamalardaki
sınırlamalar bağlamında tartışılmıştır. Sunulan kapsamlı değerlendirme, geleneksel görüntü işleme
yöntemlerinin temel matematiksel ve algoritmik yapısını ve modern yapay zekâ destekli tekniklerin
sağladığı dönüşümü bütüncül bir çerçevede ele almakta, gelecekteki araştırmalar ve uygulamalar için
hibrit modellerin entegrasyonu, çok modlu veri analizi ve gerçek zamanlı sistemler gibi olası gelişim
yönelimlerine ışık tutmaktadır.

Use of Artificial Neural Networks and Autoregressive Models for Epilepsy Detection from EEG Signals

Ferda Bozkurt — Wed, 03 Dec 2025 00:00:00 +0100

This study aimed to automatically distinguish epileptic patients from healthy individuals using
EEG signals using Electroencephalography (EEG) datasets provided by the University of Bonn. Various
autoregressive (AR) model-based methods were used for feature extraction, and a feed-forward
backpropagation artificial neural network (ANN) was applied for classification. Classification results were
evaluated using various metrics. The study demonstrated that the proposed approach can distinguish
epileptic and healthy signals with high accuracy.

IoT-Based Systems for Energy Efficiency in Smart Buildings: Present, Trends and Sustainability Perspective

Mehmet Yagiz Orhan, Aytac Ugur Yerden — Wed, 03 Dec 2025 00:00:00 +0100

This paper examines Internet of Things (IoT)-based systems developed to improve energy
efficiency in smart buildings from a sustainability perspective. Considering that buildings account for 35
40% of global energy consumption, digital transformation in this sector plays a pivotal role in achieving
Sustainable Development Goals, particularly SDG-7 (affordable and clean energy) and SDG-11
(sustainable cities and communities). The study systematically evaluates the growing role of IoT
technologies in energy management by analyzing the energy-saving potential of applications such as
HVAC control, lighting automation, solar panel monitoring, and occupancy sensors. Findings from the
literature suggest that these systems offer 15–30% energy savings, although implementation costs and
limited user adaptation remain key barriers to widespread adoption. Furthermore, the performance and
applicability of communication protocols such as ZigBee, LoRa, Wi-Fi, and MQTT are comparatively
discussed. Drawing on over 15 peer-reviewed studies published between 2020 and 2025, the paper
identifies technical and policy-related challenges such as high installation costs, data security concerns,
interoperability issues, and the lack of standard infrastructure. Additionally, it examines the current status
of policy frameworks such as the European Green Deal and Turkey's BEP-TR in relation to IoT
integration. Ultimately, the paper suggests increasing government incentives, developing technical
standards, promoting open-source solutions, and enhancing user awareness to facilitate the broader
adoption of IoT systems in smart buildings. These strategies are expected to significantly support both
environmental sustainability and economic efficiency in the context of global energy transition.

Sign Language Recognition for Vehicle Control: A Deep Learning and Arduino Approach

Bashar ALHAJAHMAD, Musa Ataş, İsa Ataş — Wed, 03 Dec 2025 00:00:00 +0100

This study presents a novel sign language recognition system designed for vehicle control,
leveraging deep learning and Arduino technology. As traditional human-computer interaction methods
evolve, gesture recognition has emerged as an intuitive alternative, particularly beneficial for individuals
with hearing disabilities. With over 466 million people affected by hearing loss globally, effective
communication tools are increasingly vital. The proposed system utilizes a pre-trained Convolutional
Neural Network (CNN), specifically the VGG19 model, which was fine-tuned on a dataset comprising
4,769 images of hand gestures. The integration of image processing techniques with Arduino enables
real-time gesture recognition and control of vehicle functionalities. The results demonstrate an impressive
accuracy of 99.79% for gesture recognition, validating the effectiveness of deep learning methodologies
in this domain. This approach not only enhances accessibility for users with hearing impairments but also
paves the way for innovative applications in smart environments, interactive gaming, and assistive
technologies.

Nar Kabuğu Ekstraktı İçeren pH-Duyarlı Biyoaktif Yenilebilir Film Tasarımı ve Ekmek Ürünlerinde Raf Ömrü İzleme Uygulamaları ile İlgili Bir Araştırma

Ayşe Nur TEMİZKAN, Mehmet Onurhan GÜCÜŞ, Süleyman GÖKMEN — Wed, 03 Dec 2025 00:00:00 +0100

Bu derleme, nar kabuğu ekstraktı kullanılarak geliştirilebilecek pH-duyarlı yenilebilir film-tablet
sistemlerinin düşük nemli gıda ürünlerinde potansiyel kullanımını değerlendirmeyi amaçlamaktadır.
Fenolik bileşikler bakımından zengin olan nar kabuğunun, güçlü antioksidan ve antimikrobiyal etkileri
sayesinde film matrisine biyoaktif özellik kazandırabileceği literatürde belirtilmektedir. Bu doğrultuda,
film-tablet formunun Wasa tipi ekmek gibi düşük nemli gıdaların yüzeyine uygulanmasının, yüzeysel
mikrobiyal bozulmayı azaltma ve pH değişimine dayalı renk dönüşümü sayesinde ürün tazeliğinin görsel
olarak izlenmesini sağlama potansiyeli bulunmaktadır. Doğal atık materyallerin değerlendirilmesi, çevresel
sürdürülebilirlik ve ekonomik katkı açısından önemli avantajlar sunmaktadır. Bu bağlamda nar kabuğu
ekstraktı içeren yenilebilir film-tablet sistemleri, hem gıda güvenliğini destekleme hem de akıllı ambalaj
teknolojilerinin geliştirilmesine katkı sağlayabilecek yenilikçi bir yaklaşım olarak değerlendirilmektedir.
Literatürde düşük nemli gıdalara yönelik akıllı ambalaj uygulamalarına ilişkin çalışmaların sınırlı olması,
bu tür sistemlerin gelecekteki araştırmalar için önemli bir fırsat sunduğunu göstermektedir.
Genel olarak bu önerilen sistem, hem aktif koruma sağlayarak raf ömrünü uzatma potansiyeline sahiptir
hem de renk değişimi yoluyla tüketiciye ürün tazeliği hakkında bilgi sunarak gıda israfının azaltılmasına
katkıda bulunabilir.

Endüstri 4.0’da Yapay Zekâ Destekli Öngörücü Bakım

Ferda Bozkurt — Wed, 03 Dec 2025 00:00:00 +0100

Bu çalışma, Endüstri 4.0 kapsamında yapay zekâ destekli öngörücü bakım (PdM - Predictive
Maintenance) uygulamalarını ve literatürdeki güncel yöntemleri incelemektedir. Sensör teknolojileri, veri
işleme yöntemleri, makine öğrenimi, derin öğrenme ve pekiştirmeli öğrenme yaklaşımları, farklı
endüstriyel sektörlerde PdM’in uygulanabilirliğini ve etkilerini ortaya koymak amacıyla literatür
taramasıyla değerlendirilmiştir. Bulgular, PdM’in ekipman güvenilirliğini artırmada, duruş sürelerini ve
bakım maliyetlerini azaltmada etkili olduğunu göstermektedir.

Polietilen Tabanlı Sistemlerde Sıcaklık ve Karışım Oranının Çentikli İzod Darbe Direncine Etkisi

Erkin Akdoğan — Wed, 03 Dec 2025 00:00:00 +0100

Bu çalışmada, yüksek yoğunluklu polietilen (HDPE) ve düşük yoğunluklu polietilen (LDPE) ile
bunların karışımının 0°C ve 23°C koşullarında çentikli İzod darbe direnci üzerindeki etkileri deneysel
olarak incelenmektedir. Polietilenler, termoplastik polimerler ailesinde yaygın kullanılan malzemeler
olup, moleküler yapılarındaki farklılıklar nedeniyle mekanik özellikleri büyük ölçüde değişkenlik
göstermektedir. HDPE, yüksek kristal oranına sahip olması nedeniyle sertlik ve mukavemetiyle öne
çıkarken, çok düşük sıcaklıklarda gevrekleşme eğilimi gösterebilmektedir. LDPE ise daha dallı bir yapıya
ve düşük kristal oranına sahip olması sayesinde esneklik ve darbe tokluğu sunmaktadır. Bu iki polimerin
karıştırılmasıyla, her iki malzemenin avantajlı özelliklerini bir araya getiren ve mekanik performansı
optimize edilmiş kompozit malzemelerin geliştirilmesi hedeflenmektedir. Araştırmada, LDPE matrisine
kütlece %50 oranında HDPE ilave edilerek %50 LDPE + %50 HDPE (L1H1) karışımı hazırlanmıştır.
Karışım, mekanik bir karıştırıcı kullanılarak homojen hale getirilmiş, ardından nemi gidermek için
kurutma işlemine tabi tutulmuştur. Numuneler, tek vidalı bir enjeksiyon makinesi kullanılarak standart
test numunesi haline getirilmiştir. Darbe dayanım testleri, ASTM D 256 standardına uygun olarak, Izod
darbe test cihazı ile 0°C ve 23°C sıcaklıklarda gerçekleştirilmiştir. Elde edilen bulgular, sıcaklık ve
polimer türünün darbe direnci üzerinde belirleyici olduğunu göstermiştir. LDPE'nin 23°C'deki verileri
mevcut olmamakla birlikte, literatürle uyumlu olarak LDPE'nin bu sıcaklıkta da HDPE'ye kıyasla daha
düşük darbe direncine sahip olduğu ancak daha elastik bir davranış sergilediği ifade edilmektedir.
HDPE'nin kristal yapısı, düşük sıcaklıklarda daha iyi darbe direnci sağlarken, LDPE'nin amorf yapısı
kırılganlık eğilimini artırmaktadır. L1H1 karışımı ise her iki polimerin özelliklerini dengeleyerek orta
düzeyde bir performans sergilemiştir. Sıcaklık artışının, polimer zincirlerinin hareketliliğini artırarak
enerji absorbsiyon kapasitesini yükselttiği ve darbe direncini olumlu yönde etkilediği gözlemlenmiştir. Bu
çalışma, polimer karışımlarının sıcaklık bağımlı mekanik davranışını nicel olarak ortaya koyması ve
mühendislik uygulamalarında optimize edilmiş malzeme seçimine katkı sağlaması açısından önem
taşımaktadır.

GAN Tabanlı Süper Çözünürlük ile İyileştirilmiş Pnömoni Tespiti

Elif Işılay ÜNLÜ TATAR, Ahmet ÇINAR — Wed, 03 Dec 2025 00:00:00 +0100

Bu çalışmada, düşük çözünürlüklü göğüs röntgeni görüntülerinin tanısal kalitesini artırmak ve
pnömoni tespit performansını iyileştirmek amacıyla GAN tabanlı bir süper çözünürlük ve derin öğrenme
tabanlı sınıflandırma mimarisinden oluşan hibrit bir framework önerilmektedir. Süper çözünürlük
modülü, RRDB tabanlı bir jeneratör ve PatchGAN ayrımcısından oluşmakta olup, düşük çözünürlüklü
(LR) görüntülerden yüksek çözünürlüklü (SR) görüntüler üretmektedir. Üretilen SR görüntüler daha
sonra pnömoni tespiti için ResNet18 tabanlı ikili sınıflandırıcıya girdi olarak verilmiştir. Modelin
performansı, Kaggle Chest X-Ray Pneumonia veri seti kullanılarak değerlendirilmiştir. Elde edilen
deneysel sonuçlara göre önerilen SR modülü, doğrulama kümesinde 31.2 dB PSNR ve 0.864 SSIM
değerlerine ulaşarak yüksek kalitede görüntü geri kazanımı sağlamıştır. SR çıktılarıyla eğitilen
sınıflandırıcı ise doğrulama kümesinde %96.67 doğruluk elde etmiş ve SR’nin tanısal performans
üzerinde anlamlı bir katkı sağladığı görülmüştür. Bulgular, GAN tabanlı süper çözünürlük ile
sınıflandırmanın birlikte kullanımının, düşük çözünürlüklü klinik görüntüler üzerinde hem görüntü
kalitesini hem de tanı doğruluğunu artıran etkili bir yaklaşım sunduğunu göstermektedir.

A Sumudu Transform Approach to Linear and Nonlinear Volterra Integral Equations

Emel Biçer — Wed, 03 Dec 2025 00:00:00 +0100

This study investigates the effectiveness of the Sumudu transform method in obtaining
analytical solutions for Volterra integral equations of the second kind. Owing to its unique structural
properties particularly its ability to preserve the original function domain and simplify convolution-type
kernels the Sumudu transform provides a powerful alternative to classical integral transform techniques
such as Laplace and Fourier transforms. In this work, we present a systematic framework for applying the
Sumudu transform to Volterra equations with various kernel structures, including separable, convolution,
and weakly singular kernels. The transformed equations are reduced to solvable algebraic forms, enabling
straightforward inversion to obtain closed-form or semi-analytical solutions. Several illustrative examples
are provided to demonstrate the accuracy, simplicity, and computational efficiency of the proposed
approach. The results confirm that the Sumudu transform is an effective and reliable tool for solving a
wide class of Volterra integral equations, offering potential advantages for applications in mathematical
physics, engineering, and applied sciences.

Yazılım Mühendisliğinde Kodlama Ajanı Devrimi

Musa Ataş, Bashar Alhajahmad, İsa Ataş — Wed, 03 Dec 2025 00:00:00 +0100

Yazılım geliştirme pratiği, deterministik ve kural tabanlı CASE araçlarının hakimiyetinden, niyet
odaklı (intent-based) ve stokastik çıktılar üreten Agentic Coding sistemlerinin şekillendirdiği yeni bir
paradigmaya dönüşmektedir. Bu çalışma, söz konusu süreci teknik bir verimlilik optimizasyonu olarak
değil, yazılım mühendisliği pratiğini ve profesyonel kimliğini temelden yeniden yapılandıran ontolojik bir
kırılma olarak ele almaktadır. Actor-Network Theory (ANT) ve Dağıtık Biliş (Distributed Cognition)
teorilerini sentezleyen bu araştırma, insan geliştiriciler ve AI ajanları arasındaki etkileşimi, statik bir araç
kullanıcı ilişkisinden öte, bilişsel süreçlerin paylaşıldığı "Ortak Bilişsel Sistemler" (Joint Cognitive
Systems) olarak analiz etmektedir. Makale, bu etkileşimin doğasını sınıflandırmak üzere otonomiden
hibrit orkestrasyona uzanan dört katmanlı bir "İş Birliği Spektrumu" sunmaktadır. AI çıktılarının
olasılıksal doğasından kaynaklanan epistemik belirsizlik, "halüsinasyon" riski ve kurumsal düzeyde
sorumluluk difüzyonu gibi kritik problemler tartışılmaktadır. Ayrıca, yazılım ekiplerinin bu yeni aktörlere
adaptasyonunu modelleyen, zamansal dinamiklere dayalı özgün bir Olgunluk Çerçevesi (Maturity
Framework) önerilmektedir. Çalışmanın temel tezi, sürdürülebilir ve verimli bir insan-AI simbiyozunun;
kod üzerindeki geleneksel "yazarlık" (authorship) kavramının yeniden müzakere edilmesine, şeffaf ve
doğrulanabilir güven protokollerinin tesis edilmesine ve geliştiricilerin teknik ustalık ile bilgelik
arasındaki dengeyi yeniden kurmasına bağlı olduğu ile ilintilidir.

An In Silico Approach Study: The Predicted Toxicity Profile of Diazinon

Nazli Turkten, Yunus Karatas, Yelda Yalcin Gurkan, Simal Kurumoglu — Wed, 03 Dec 2025 00:00:00 +0100

-The increased use of pesticides in the agricultural sector to meet global food demand, driven by
population growth. The widespread application of organophosphorus pesticides in crop production poses a
significant environmental problem related to the accumulation of residues in soil and surface water. The
presence of these organophosphorus residues can alter soil nutrient diversity, which is influential on
microorganisms responsible for soil fertility. Moreover, this agricultural-originated problem can cause toxic
effects to the ecosystem and human health. For that reason, the toxicity profile of diazinon (DZ), an
organophosphorus pesticide, was assessed by using two different in silico approaches. The predicted
toxicity of DZ was evaluated by using ProTox 3.0 and TEST programs. The results indicated that the DZ
pesticide exhibited toxic signs associated with the calculated lethal dose 50 (LD50) values.

Interface Model for Blind and Visually Impaired for Albanian language

Lindita Ademi, Valbon Ademi — Wed, 03 Dec 2025 00:00:00 +0100

Being able to read text, find out information and know about the latest news has always been a
challenge for those who cannot access the printed version, such as the blind and visually-impaired.
The advent of internet and advancements in User Interfaces (UIs) as well as in assistive technologies make this
challenge increasingly feasible. A lot of effort has gone into making software applications and web pages more
accessible to the blind and visually impaired students. There are few Windows based stand-alone text-to-speech
screen reader applications that provide application-specific speech output through script files written
specifically for them. All these screen readers usually cover widely speaking languages. Albanian language is
not covered with no one of the screen readers.
In this paper a new user-interface model for Albanian speaking blind and visually impaired is
presented. The new model is built up on the research results and testing with different screen readers and
different groups of blind and visually impaired Albanian speaking students, the screen readers analysis, text
to-speech technologies analysis, specially developed text-to-speech generator from Albanian texts, the
mathematical analysis done on general Albanian texts as well as the basic principles of lexicons creations. It
has been tested to blind and visually impaired science students. The new user-interface model is a very useful
tool for Albanian-speaking blind and visually impaired science students giving them an equal studying
opportunity as the other students.

Integrating exploratory GAMs into the design of Bayesian BYM2 spatio-temporal models for small-country disease mapping

Fabiana Çullhaj, Henri Buçka, Raimonda Dervishi — Fri, 05 Dec 2025 00:00:00 +0100

Statistical disease mapping is central to public health surveillance, yet small-country settings with few regions and relatively short time series pose challenges for flexible yet identifiable spatio-temporal models. This article proposes a practical two-stage workflow that integrates generalized additive models (GAMs) with Bayesian spatio-temporal disease mapping, tailored to surveillance of acute respiratory infections, enteric infections and similar outcomes. In the first stage, we fit exploratory GAMs with smooth terms for calendar time, seasonality, spatial location and key covariates. Partial-effect plots and residual diagnostics are used to characterise non-linear relationships, seasonal patterns and spatial gradients, and to identify residual dependence that may require explicit modelling. In the second stage, we formalize these insights into a Bayesian latent Gaussian model fitted with INLA, combining a BYM2 spatial component, random-walk (RW1/RW2) temporal effects, optional cyclic seasonal terms and, where warranted, parsimonious spatio-temporal interactions. Penalised complexity (PC) priors regularise spatial and temporal hyperparameters, encouraging simple structure while allowing genuine variation to emerge. Lessons from small-country applications highlight the importance of balancing model complexity and identifiability, exploiting GAMs as a design tool rather than an endpoint, and using PC priors to stabilise inference when information is limited. The workflow offers a concrete template for combining exploratory flexibility and hierarchical regularisation in disease mapping for data-constrained surveillance systems, particularly in settings with few administrative regions and relatively short time series.

The Technological Development, Applications and Social Impact of Chatbots in the Modern Digital Environment

József Udvaros, Veronika Gabaľová — Fri, 05 Dec 2025 00:00:00 +0100

The rapid development of artificial intelligence (AI) has fundamentally transformed human-machine interaction. Chatbots have become one of the most widely used AI technologies, effectively supporting customer service, education, finance and healthcare processes. This paper reviews the technological development of chatbots, their types, their operation based on natural language processing, and examines their application through practical case studies. Special attention is paid to ethical, data security and cultural challenges, as well as future development directions. The paper builds on current scientific literature, with a special focus on the new possibilities of generative AI and large language models (LLMs).

The Use of Artificial Intelligence in Education: Trends, Challenges and Innovative Approaches

József Udvaros, Veronika Gabaľová — Fri, 05 Dec 2025 00:00:00 +0100

The digital transformation of education is increasingly driven by artificial intelligence (AI), offering significant opportunities to enhance learning experiences through personalized instruction, interactive content creation, and more inclusive approaches to education. AI technologies enable adaptive learning systems that respond to individual students’ needs, providing tailored feedback, learning pathways, and resources that support diverse learning styles and abilities. This article examines current applications of AI in schools, emphasizing practical insights from DigiEduHack Bratislava, an international hackathon that encourages the development of innovative educational solutions and promotes collaboration among students, educators, and technology experts. Additionally, the use of Genially.com is highlighted as a tool for creating interactive and AI-enhanced learning materials that foster student engagement and creativity. The paper further explores trends in AI integration within classrooms, discussing pedagogical benefits such as improved motivation, active participation, and enhanced critical thinking skills. It also addresses challenges, including ethical and legal considerations, data privacy, technological limitations, and the need for teacher training to ensure effective and responsible implementation. Finally, the article provides actionable recommendations for educators, schools, and policymakers, focusing on strategies to successfully incorporate AI tools into educational practice while maintaining a human-centred approach. By combining technological innovation with pedagogical principles, AI has the potential to improve learning outcomes, encourage creativity, and promote inclusive education, ensuring that digital transformation supports both students and teachers in meaningful ways.

Aligning the Western Balkans with the EU’s Farm to Fork Strategy: Impact in Sustainable Agricultural Transformation

Eva Hyna — Fri, 05 Dec 2025 00:00:00 +0100

The European Union’s “Farm to Fork” (F2F) Strategy outlines a comprehensive roadmap for transforming food systems to align with sustainability goals. By addressing every stage of the food chain, from production and processing to consumption and waste, the strategy aims to create a food system that is environmentally responsible, economically viable, and socially inclusive. For the Western Balkan countries, which are pursuing long-term integration into the EU, aligning with the F2F Strategy is not just a regulatory requirement but also a crucial step toward securing rural livelihoods, enhancing food security, modernizing agricultural practices, and mitigating environmental and climate risks. This paper explores the strategic importance of the F2F Strategy for the Western Balkans and assesses the region’s agricultural structures in light of EU sustainability objectives. It identifies major opportunities, such as increased access to EU funding instruments, improved productivity, and strengthened rural development, but also highlights several structural and institutional challenges. These include fragmented farm structures, limited technological capacity, weak enforcement of food safety standards, and high vulnerability to climate change. The analysis emphasizes the need for integrated governance models, targeted financial mechanisms, enhanced advisory services, and market-driven consumption incentives. These are essential to ensure that alignment with the F2F agenda becomes a catalyst for transformation rather than a burden. The study offers practical recommendations for facilitating a fair, effective, and sustainable agricultural transition in the Western Balkans.

Magnetic Displacement Sensor Based Elevator Weight Measurement

Sat, 06 Dec 2025 00:00:00 +0100

Elevator is famous and common way for vertical transporting. Today, elevator traffic management systems are so important to reduce energy and time losses. In order to obtain time and prevent energy losses, elevator weight must be measure and control strategies must regard the weight. There are two classical methods to measure the weight. one of them works as one of and the other is based on real load cells. In this study a novel method is offered to measure elevator weight regarding easy installing and calibrating. In this study, high resolution magnetic displacement sensor is designed and characterized for elevator application. Magnetic displacement sensor is sensitive to magnetic field changing. In this study, 1cm distance are calibrated as 1000kg and output generates 0-10V voltage related to distance. Inside the sensor includes FIR sensor to reduces fluctuating. Because elevator includes rubber spring to reduce mechanical noise and shaking.

Dealing with Weight Measurement Problems in Elevator Applications

Sat, 06 Dec 2025 00:00:00 +0100

The Elevators are found in many areas of our lives. Elevators hold a special place in high-rise buildings, especially as vehicles for transporting people and cargo. Depending on the need, it's possible to find more than one type of elevator in the same building. Freight elevators and passenger elevators can vary in type depending on their speed and carrying capacity. However, one of the most important parameters in all applications is the load capacity of the elevator. Measuring the weight carried by an elevator is difficult due to factors such as its motion and exposure to vibration. This study addresses these weight measurement challenges and explores the fundamentals of designing a sensor resistant to these challenges. Applications are specifically targeted at vertically moving elevators. Accurately determining whether an elevator is empty, fully loaded, or overloaded is crucial for both the health of elevator traffic and its operational safety. Accurately measuring this parameter will prevent time, energy, and economic losses, thus extending the elevator's effective service life. In this study, gauss-based sensor is examined as weigh monitoring and evaluation to manage the elevator control.

Remotely Addressable Buttons with Can-Bus Data Path in Elevator Applications

Neşe ÖZEN DEMİR, Burak DEMİR , Murat TOPUZ, Bayram AKDEMİR — Sat, 06 Dec 2025 00:00:00 +0100

CAN-Bus and its derivatives are widely used in many applications. In this study, CAN-Bus communication has been specifically developed for elevator buttons. Traditional elevator buttons have numerous cables to communicate with the control panel. However, buttons only help us call the elevator. Old-type buttons in high-rise buildings have a large number of cables. A large number of cables frequently cause malfunctions. CAN-Bus is a solution to reduce the number of cables. In contrast, faulty buttons based on CAN-Bus communication are irreparable due to software programming. This study presents a new button style that can be dynamically readdressed remotely and locally. Reducing the number of buttons provides significant efficiency in sales and profitability in the market. Data transmission based on the CAN-Bus also offers convenience in addressing and interactive solutions. In case of a malfunction, easy access to the button, reduced wiring, energy transmission on high floors, and cost savings are achieved. This system aims to design a low-cost and modular system that communicates via the CAN Bus, displaying information from multiple data sources (speed, temperature, fault codes, etc.) on a dot matrix display in real-time. Furthermore, CAN Bus communication dot matrix indicators provide ease of use in elevator systems for floor information, direction, error messages, and faults and current status occurring in the systems. It provides data transmission to passengers both inside and outside the cabin.

Yenilenebilir Enerji Arayüzlerinde Güç Yarı İletkenlerinin Ömür Analizi: Görev Profiline Dayalı Elektro-Termal Modelleme

Turan ÇAKIL — Sun, 07 Dec 2025 00:00:00 +0100

Yenilenebilir enerji kaynaklarının (YEK) artan penetrasyonu, güç elektroniği dönüştürücülerinin
güvenilirliğini şebeke kararlılığı açısından kritik bir parametre haline getirmiştir. Özellikle rüzgâr ve
fotovoltaik sistemlerin stokastik (rastgele) güç üretim profilleri, IGBT güç modülleri üzerinde geleneksel
endüstriyel uygulamalardan farklı termomekanik stresler oluşturmaktadır. Bu çalışma, gerçek saha
verilerine dayalı görev profillerini kullanarak, yarı iletkenlerdeki termal döngüleri ve buna bağlı ömür
tüketimini analiz eden entegre bir model sunmaktadır. Foster ısıl ağ modeli ile elektriksel kayıpların
dinamik dönüşümü sağlanmış, Rainflow sayımı algoritması ve Miner kuralı ile kümülatif hasar
hesaplanmıştır. Sonuçlar, düşük frekanslı termal döngülerin, toplam hasarın %70'inden fazlasını
oluşturduğunu ve soğutma stratejilerinin sadece maksimum sıcaklığa (Tj, max) değil, sıcaklık salınım
genliğine (ΔTj) odaklanması gerektiğini göstermektedir.

Endüstriyel Kalite Kontrolünde Mermer Anomali Tespit Teknikleri: Kapsamlı Bir Derleme

Musa Ataş, Bashar Alhajahmad, İsa Ataş — Sun, 07 Dec 2025 00:00:00 +0100

Mermer üretim endüstrisinde yüzey kusurlarının tespiti, ürün kalitesi ve ekonomik değer açısından
kritik öneme sahiptir. Geleneksel manuel muayene yöntemleri subjektif, zaman alıcı ve hataya açık
olduğundan, otomatik kusur tespit sistemlerine olan ihtiyaç gün geçtikçe artmaktadır. Bu derleme
çalışması, mermer yüzeylerindeki çatlak, renk değişimi, leke ve doku anomalilerinin tespitinde kullanılan
bilgisayarlı görü ve makine öğrenmesi tekniklerini derleme sistematiği penceresinde ele almaktadır.
Klasik görüntü işleme yöntemlerinden (Gabor filtreleri, LBP, wavelet dönüşümleri) modern derin
öğrenme mimarilerine (U-Net, FPN, YOLO, DeepLabv3+) kadar geniş bir yelpazede teknikler
incelenmiştir. RGB ve termal görüntüleme modalitelerinin füzyonu, özellikle FPN tabanlı modellerde
mIoU değerlerini %87'ye kadar çıkarabilmektedir. Denetimli segmentasyon yöntemleri yüksek
lokalizasyon performansı sunarken, otoenkoder ve metrik öğrenme tabanlı denetimsiz yaklaşımlar etiket
gereksinimini azaltmakta ve dengesiz veri setlerinde daha kararlı sonuçlar vermektedir. YOLOv8-Seg
gibi domain-optimize edilmiş mimariler, endüstriyel uygulamalarda mAP@0.5 değerlerini 0.856
seviyelerine çıkarabilmektedir. Bu çalışmada ayrıca 2020-2025 yılları arasındaki güncel gelişmeler,
performans karşılaştırmaları, uygulama zorlukları ve gelecek araştırma yönelimleri detaylı olarak
tartışılmaktadır. Yapılan kapsamlı araştırmalar neticesinde, multimodal görüntü füzyonu, hafif
segmentasyon mimarileri ve denetimsiz öğrenme yaklaşımlarının entegrasyonu, gerçek zamanlı
endüstriyel uygulamalar için en fazla umut verici teknikler olarak bir adım öne çıkmaktadır.

Sürdürülebilir Kentleşme Üzerine Değerlendirmeler

Özlem Candan HERGÜL, Serdar KASAP — Sun, 07 Dec 2025 00:00:00 +0100

Sürdürülebilirlik kavramı, ekonomik ve teknolojik gelişmelerin beraberinde getirdiği çevresel
sorunlara karşı bir çözüm arayışı olarak ortaya çıkmıştır. Sürdürülebilir kent kavramı, uzun vadeli bir
süreci kapsamakta ve kentlerin çevresel, ekonomik ve sosyal anlamda dengeli bir şekilde gelişimini
hedeflemektedir. Bu çalışma kapsamında, sürdürülebilir kent kavramı irdelenmekte; kentlerin
sürdürülebilirliğinin sağlanmasının çevre sorunları bağlamında değerlendirilmesi ve insanların yaşam
kalitelerinin artırılmasına yönelik gereksinimler incelenmektedir. Sürdürülebilir kentleşme, yalnızca
çevresel faktörleri değil, aynı zamanda toplumsal ve ekonomik uyumu da göz önünde bulundurmayı
gerektiren bir kavramdır. Kentlerin büyüme süreçleri, modern yaşamın gerekliliklerini karşılamakla
birlikte ekolojik dengeyi koruyan politikalarla desteklenmek zorundadır. Çalışmada, örnek kentler
incelenerek sürdürülebilirlik yaklaşımına farklı perspektiflerden bir bakış sunmak hedeflenmektedir.

UV-Vis-NIR Photodetector Produced Using Lavandula angustifolia Extract: A Green Approach

Mehmet Hakan ÇOLPAN — Sun, 07 Dec 2025 00:00:00 +0100

In this study, the usability of Lavandula angustifolia (lavender), a natural plant extract, as an
active material in the production of photodetectors was investigated. An environmentally friendly and
self-powered photodetector capable of broad-band detection in the UV-Vis-NIR (ultraviolet-visible-near
infrared) spectral range was successfully produced with a simple and low-cost method. The basic
performance parameters of the photodetector, namely photosensitivity (Iph), photosensitivity (K),
responsivity (R) and specific detectability (D*), were investigated in detail at different wavelengths. The
obtained data show that the device exhibits high performance especially in the range of 351–1000 nm and
reaches maximum values at 1000 nm (Iph=1.58×10−5 A, R=0.1 A/W, D*=1.82×1010 Jones). Significant
decreases were observed in these parameters at wavelengths above 1000 nm. These results demonstrate
that lavender extract-based natural material offers a promising photodetector platform for applications
from UV to near-infrared.

In Silico Toxicity Assessment of Chalcone Derivatives—Derricin, Lonchocarpin, and Isocordoin—from Lonchocarpus cultratus

Nergiz Cengiz, Simal Kurumoglu — Sun, 07 Dec 2025 00:00:00 +0100

Chalcones, widely occurring plant secondary metabolites, constitute a structural subgroup of the
flavonoid family. They have attracted considerable interest due to their diverse biological activities,
including antibacterial, anticancer, antimalarial, antileishmanial, anti-inflammatory, and antidiabetic
effects. The protozoan parasites Trypanosoma brucei and Trypanosoma cruzi (T. cruzi) are the primary
causes of African trypanosomiasis and Chagas disease, affecting approximately eight million people
worldwide. Current therapeutic options for these diseases are limited and often associated with significant
toxicity, underscoring the need for safer and more effective alternatives.
Lonchocarpus cultratus is a tree native to tropical regions, and previous research suggests that extracts
derived from its roots can be used in the treatment of T. cruzi infections (Bortoluzzi et al., 2021). Among
the chalcones identified in these root extracts, derricin, lonchocarpin, and isocordoin have demonstrated
notably strong anti-T. cruzi activity. In this study, the toxicity profiles of these three chalcone compounds—
considered promising candidates for T. cruzi treatment—were evaluated using in silico methods with
ProTox 3.0.

Design, Implementation, and Performance Analysis of a Bluetooth-Based Smart Home Appliance Control System Using Arduino Microcontroller

Thu, 11 Dec 2025 00:00:00 +0100

This study presents the design, implementation, and performance analysis of a Bluetooth-based
smart home appliance control system using an Arduino microcontroller platform. The system integrates
an Arduino Uno, an HC-05 Bluetooth module, and relay circuits to enable wireless control of home
electrical appliances through a smartphone application. The primary goal of the system is to enhance user
convenience, safety, and energy efficiency by replacing traditional manual switches with a reliable, low
cost, and user-friendly automation solution. Experimental testing was conducted on multiple household
appliances, including lights and fans, to evaluate key performance parameters such as response time,
communication range, reliability, and power consumption. The results demonstrated an average response
time between 0.7 and 1.2 seconds, a stable Bluetooth communication range of approximately 10 meters,
and a command reliability rate of 97–100%, with minimal power consumption throughout operation. The
modular structure of the system supports scalability and easy integration with future Internet of Things
(IoT) technologies, expanding its potential for advanced home automation applications. Overall, the
research confirms that a Bluetooth-controlled Arduino-based system provides an effective, energy
efficient, and cost-efficient solution for modern smart home environments.