5794 research outputs found
Sort by
Hybrid Whale and Artificial Rabbit Optimization for Efficient Multi‑Objective Sensor Deployment in Complex IoT Networks
No full textThis paper presents a novel hybrid metaheuristic algorithm, combining Whale Optimization Algorithm (WOA) and Artificial
Rabbits Optimization (ARO), to solve the multi-objective sensor node placement problem in dynamic and obstacle-rich
Internet of Things (IoT) environments. The proposed WOA-ARO algorithm aims to maximize coverage, minimize energy
consumption, and reduce redundancy while maintaining robust network connectivity. Leveraging WOA’s strong global search
capabilities alongside ARO’s efficient local refinement, the hybrid method balances exploration and exploitation effectively.
Extensive simulations conducted on real-world maps with 50 sensor nodes demonstrate that WOA-ARO achieves an average
coverage rate of 95.00% with a remaining energy of 88.31%, outperforming competing algorithms such as EFFSA, MAOA,
and GA-PSO. Additionally, WOA-ARO achieves the lowest redundancy value of 1.2142, indicating efficient resource utilization.
Although its runtime is marginally higher than some methods, the superior solution quality and energy efficiency
affirm WOA-ARO as a highly effective approach for optimal sensor deployment in complex IoT scenarios
AI-Powered Prediction of Dental Space Maintainer Needs Using X-Ray Imaging: A CNN-Based Approach for Pediatric Dentistry
No full textSpace maintainers (SMs) are essential for preserving dental arch integrity after
premature tooth loss. This study aimed to develop a deep learning model to predict the
necessity of SMs and identify specific teeth requiring intervention. A dataset of 400 dental
X-rays was preprocessed to standardize image dimensions and convert them into numerical
representations for machine learning. The dataset was divided into training (80%) and
testing (20%) subsets. A Convolutional Neural Network (CNN) was designed with multiple
convolutional and pooling layers, followed by fully connected layers for binary classification.
The model was trained using 30 epochs and evaluated with accuracy, precision,
recall, F1-score, ROC AUC, and MCC. The CNN achieved 94% accuracy, with a precision
of 0.93 for Class 0 (no SM needed) and 0.95 for Class 1 (SM needed). The ROC AUC
was 0.94, and the MCC was 0.875, indicating strong reliability. When tested on 86 X-ray
images, the model successfully identified specific teeth (showing teeth number) requiring
SMs, with minimal errors. These results suggest that the proposed AI model provides
high-performance predictions for SM necessity, offering a valuable decision-support tool
for pediatric dentistry
Comparison of cloud-based ai-powered and user-controlled software in terms of 3D visualization for interior design
No full textİç mekan tasarımı, estetik, işlevsellik ve kullanıcı deneyimini bir arada ele alan disiplinlerarası bir çalışma alanıdır. Tarih boyunca tasarımın daha etkili bir şekilde ifade edilmesi ve anlaşılması amacıyla kullanılan görselleştirme yöntemleri, teknolojinin ilerlemesiyle birlikte büyük bir dönüşüm geçirmiştir. Geleneksel analog yöntemler, eskiz, perspektif ve maket gibi tekniklerle tasarımın temel unsurlarını ortaya koyarken; dijital teknolojiler, 2D ve 3D modelleme, sanal gerçeklik ve artırılmış gerçeklik gibi araçlarla bu süreci daha dinamik ve etkili hale getirmiştir. Son yıllarda yapay zeka teknolojilerinin görselleştirme süreçlerinde kullanımı, tasarım dünyasında önemli yenilikler oluşturmuştur. Yapay zeka, veri işleme ve karar alma süreçlerini hızlandırarak hem zamandan tasarruf sağlamış hem de işlem gücünü optimize etmiştir. Bu tez çalışmasında, kullanıcı kontrollü render yöntemleri ile yapay zeka destekli görselleştirme sistemleri karşılaştırılmıştır. Loft tarzı bir daire modeli üzerinden gerçekleştirilen görselleştirme süreci, her iki yöntemin performansını değerlendirmek için bir çerçeve sunmuştur. kullanıcı kontrollü yöntemlerin detay kontrolü ve profesyonel kalite sağlama açısından üstün olduğu gözlemlenirken, yapay zeka destekli sistemlerin hız, erişim kolaylığı ve işlem gücü açısından avantaj sunduğu belirlenmiştir. Bununla birlikte, yapay zeka sistemlerinin üretkenliğini sınırlayan bazı dezavantajlar içerdiği tespit edilmiştir. Araştırma kapsamında yapılan anket ve istatistiksel analizler, kullanıcıların gerçekçilik, ışıklandırma doğallığı ve malzeme detaylandırması gibi kriterlerde manuel yöntemleri daha başarılı bulduklarını; buna karşın hız ve kolaylık açısından yapay zeka sistemlerini tercih ettiklerini ortaya koymuştur. Sonuç olarak, bu iki yöntemin iç mekan tasarım süreçlerinde tamamlayıcı bir şekilde kullanılabileceği sonucuna varılmıştır. Bu çalışma, iç mimarlık öğrencileri ve profesyoneller için görselleştirme teknolojilerinin avantaj ve sınırlamalarını anlamalarına yönelik bir rehber niteliği taşımaktadır. Gelecekte, yapay zeka teknolojilerinin daha fazla özelleştirme ve üretken kontrol sunması beklenmektedir. Anahtar kelimeler: İç Mekan Tasarımı, Yapay Zeka, Fotogerçekçi Görselleştirme, Dijital Modelleme, Tasarım Süreçleri Analizi, Bulut Tabanlı Görselleştirme YazılımlarıInterior design is an interdisciplinary field that integrates aesthetics, functionality, and user experience. Throughout history, visualization methods have undergone significant transformations with the advancements in technology, aiming to express and communicate design ideas more effectively. Traditional analog methods, such as sketching, perspective drawing, and model-making, have provided foundational tools for design. Meanwhile, digital technologies, including 2D and 3D modeling, virtual reality, and augmented reality, have made the design process more dynamic and impactful. In recent years, the incorporation of artificial intelligence (AI) into visualization processes has introduced substantial innovations in the design industry. AI accelerates data processing and decision-making, enabling time savings and optimizing computational power. This thesis compares manual rendering methods with AI-supported visualization systems. A loft-style apartment model was used to evaluate the performance of both approaches within a structured framework. Manual methods were observed to excel in providing detailed control and professional quality, whereas AI-supported systems demonstrated advantages in speed, accessibility, and computational efficiency. However, it was also identified that AI systems introduce certain limitations by constraining creative control. Surveys and statistical analyses conducted during the research revealed that users found manual methods superior in criteria such as realism, lighting naturalness, and material detailing. Conversely, AI systems were preferred for their speed and ease of use. Consequently, it was concluded that these two methods could complement each other in interior design workflows. This study serves as a guide for interior design students and professionals, helping them understand the advantages and limitations of visualization technologies. In the future, AI technologies are expected to offer greater customization and enhanced creative control. Keywords: Interior Design, Aı-Supported Design, Photorealistic Visualization, Digital Modeling, Design Process Analysis, Cloud-Based Visualization Softwar
Abdullatif Karamani and his work Adâb-i Menâzil: Analysis and critical text
No full textOsmanlı tasavvuf geleneğinde "şeyh" unvanı ile tanınan Abdüllatif Ka-ramânî'nin, ahlak-nâme türünde ele aldığı Âdâb-ı Menâzil adlı eserinin tahkik ve ince-lemesini konu edinmektedir. Çalışmada Âdâb-ı Menâzil'in ahlak eserleri geleneği içindeki yeri tespit edilerek, eserin benzer metinlerdeki ortak ve ayrılan yönleri karşılaştırmalı olarak ele alınmıştır. Abdüllatif Karamânî'nin hayatı ve eserlerine dair temel bilgilere yer verilerek ahlak literatürünün ana temalarından bir olan hane kavramı, müellifin bakış açısıyla değerlendirilmiştir. Bu bağlamda hane kavramı ekonomik ve sosyolojik perspektifle ele alınıp, bireylerin hane içindeki görev ve sorumlulukları ile müellifin aile yapısına dair görüşleri incelenmiştir. Kadın, erkek ve çocukların aile içindeki konumları ayrıntılı bir biçimde tartışılarak, dönemin toplumsal yapısına ışık tutulmaya çalışılmıştır. Çalışmada müellifin eserini kaleme alırken yararlandığı hadis, tefsir, edebiyat, tıp gibi farklı ilim dallarına ait kaynaklar ayrıntılı şekilde analiz edilmiş, bu kaynakların mahiyeti, Karamânî tarafından kullanım biçimleri örnekler eşliğinde ortaya konul-muştur. Bunun yanı sıra eserin günümüze ulaşan nüshaları tespit edilerek bir araya getirilmiş, bu nüshaların tavsifleri ve seçilen örnekler üzerinden tahkik çalışması gerçekleşmiştir.This study has been conducted on the critical (tahq?q) and examination of ?d?b al-Man?zil, a work authored by Abdüllatif Karam?n? who has been known in the Ottoman Sufi tradition by the title of "shaykh." Within the scope of the study, the psition of ?d?b al-Man?zil, within the tradition of ethical treatises (akhl?q-n?me) has been identifed, and its common and divergent aspects compared to other texts of the same genre have been analyzed comparatively. Fundamental infornation regarding the life and Works of Abdüllatif Ka-ram?n? has been provided, and the concept of the household ((h?ne), which has been considered one of the main themes of ethical literatüre, hes been evaluated from Ka-ram?n?'s perspective. İn this context, the notion of the household examined from both economic and sociological perspectives, and Karam?n?'s views on the duties and responsibilities of individuals within the household and onthe family structure have been analyzed. The positions pf womwn, mwn and children within the family has been discussed in detail and efforts have been made to shed ligth on the social struc-ture of the period. Moreover, the sources that the author has utilized whil composing his work -including hadith, tafs?r, literatüre and medicine -have been analyzed in detail and the natüre of these sources and the manner in wich they have been employed by Ka-ram?n? have been illustrated with wxamples. In addition, the exant manuscripts of the work have been identified and gathered their characteristic have been and class-fied and critical edition has been prepared based on the selected exemplars
An investigation of the formal structre and socio-cultural factors influencing form in traditional Erzurum houses
No full textErzurum evleri hem mimarî hem de sosyokültürel boyutlarıyla geleneksel Türk evlerinin önemli bir temsilcisidir. Bu evlerin tasarımı, bölgenin sert iklim koşulları, topoğrafik yapısı ve günlük yaşamın ihtiyaçlarına uyum sağlayarak biçimlenmiştir. Köklü bir kültürün ürünü olan Erzurum evleri, yalnızca fiziksel varlıklarıyla değil, aynı zamanda halkın yaşam biçimini, geleneklerini ve toplumsal değerlerini yansıtan somut bir kültürel miras niteliğindedir Bu çalışmada, Erzurum evlerinin biçimsel özellikleri ve bu özelliklere etki eden sosyokültürel faktörler incelenmiştir. Çalışmanın temel amacı, Erzurum evlerinin geleneksel Türk mimarî sindeki yerini ve kültürel anlamını ortaya koymaktır. Bu bağlamda, Erzurum evleri yalnızca mimarî unsurlar olarak değil, aynı zamanda toplumsal ve kültürel bağlamın fiziksel yansımaları olarak değerlendirilmiştir. Araştırma, mimarî biçimlerin çevresel faktörler ve toplumsal dinamiklerle nasıl şekillendiğini anlamaya odaklanmıştır. Araştırmanın yöntemsel çerçevesi, tarihsel ve mekânsal analizlerin yanı sıra yazılı kaynakların ve yerinde gözlemlerin disiplinler arası bir bakış açısıyla değerlendirilmesine dayanmaktadır. Elde edilen veriler, Erzurum evlerinin özgün kimliğini anlamak ve bu kimliği daha geniş bir bağlamda incelemek amacıyla kullanılmıştır Erzurum evleri, yalnızca şehrin tarihsel ve kültürel mirasını yansıtan yapılar değil, aynı zamanda toplumsal hafızayı somutlaştıran birer belge niteliğindedir. Ancak bu yapılar, günümüzde kentleşme, işlevsel yetersizlikler ve çevresel etkiler nedeniyle yok olma riskiyle karşı karşıyadır. Bu mirasın korunması, Erzurum'un kimliğini oluşturan değerlerin sürdürülebilirliğine katkı sağlayacak ve gelecek kuşaklara aktarılarak, şehrin kültürel zenginliğini yaşatacaktır.Erzurum houses represent a significant example of traditional Turkish dwellings, distinguished by their architectural and socio-cultural dimensions. The design of these structures has evolved in response to the harsh climatic conditions of the region, its topographical features, and the functional requirements of daily life. As products of a deeply entrenched cultural heritage, Erzurum houses transcend their physical existence to serve as tangible manifestations of the community's lifestyle, traditions, and social values. This study undertakes an analysis of the formal characteristics of Erzurum houses and examines the socio-cultural factors that have influenced their development. The primary objective is to elucidate the role and cultural significance of Erzurum houses within the broader context of traditional Turkish architecture. From this perspective, the study positions these houses not merely as architectural artifacts but as physical embodiments of societal and cultural dynamics. The research thus emphasizes the interplay between environmental factors, social dynamics, and architectural forms. The methodological approach integrates historical and spatial analyses with the review of written sources and direct on-site observations, adopting an interdisciplinary framework. The data collected are employed to interpret the distinctive identity of Erzurum houses and to situate this identity within a larger historical and cultural context. Erzurum houses are more than structures that reflect the historical and cultural legacy of the city; they are also material archives of social memory. Nevertheless, these buildings face significant threats due to urbanization, functional obsolescence, and environmental challenges. The preservation of this architectural heritage is imperative for sustaining the cultural values that constitute Erzurum's identity. Such efforts will ensure the transmission of these values to future generations, thereby safeguarding the city's cultural richness and historical continuity
MOLiNAS: Multi‑Objective Lightweight Neural Architecture Search for Whole‑Slide Multi‑Class Blood Cell Segmentation
No full textBlood cell analysis plays a key role in clinical diagnosis and hematological research. The accurate identification and
quantification of different blood cell types is essential for the diagnosis of various diseases. The conventional manual
method of blood cell analysis is both laborious and time-consuming, highlighting the need for automated segmentation
techniques. In this paper, the blood cell segmentation problem is considered as a multi-class segmentation
problem to detect the different types of blood cells in a given image. Two new multi-objective lightweight neural
architecture search (NAS) algorithms (MOLiNAS) are designed to tackle the challenge of whole-slide multi-class blood
cell segmentation problems. Our approaches integrate the most advantageous aspects of different approaches to
search for the best U-shaped network architecture. The performance of our approaches is compared with lightweight
networks and NAS studies in the literature. Our best solution (MOLiNASv2_sol3) achieves an IoU of 87.33 ± 1.53%, F1
score of 91.69 ± 1.20%, Precision of 93.50 ± 1.15%, and Recall of 91.34 ± 0.01%, outperforming lightweight networks
such as EfficientNet, MobileNetv2, and MobileNetv3 across all segmentation metrics. Moreover, our approaches demonstrate
highly competitive performance by utilizing up to 7.38 times fewer FLOPs and up to 4.03 times fewer trainable
parameters than existing NAS studies while requiring only 0.07 million parameters. Additionally, ablation studies
and cross-dataset evaluations demonstrate the robustness and generalizability of our approach
Optimizing Pre-Trained Code Embeddings With Triplet Loss for Code Smell Detection
Get PDFCode embedding represents code semantics in vector form. Although code embedding-based
systems have been successfully applied to various source code analysis tasks, further research is required
to enhance code embedding for better code analysis capabilities, aiming to surpass the performance and
functionality of static code analysis tools. In addition, standard methods for improving code embedding
are essential to develop more effective embedding-based systems, similar to augmentation techniques in
the image processing domain. This study aims to create a contrastive learning-based system to explore the
potential of a generic method for enhancing code embedding for code classification tasks. A triplet lossbased
deep learning network is designed to optimize in-class similarity and increase the distance between
classes. An experimental dataset that contains code from Java, Python, and PHP programming languages and
4 different code smells is created by collecting code from open-source repositories on GitHub. We evaluate
the proposed system’s effectiveness with widely used BERT, CodeBERT, and GraphCodeBERT pretrained
models to create code embedding for the code classification task of code smell detection. Our findings
indicate that the proposed system may offer improvements in accuracy, an average of 8% and a maximum of
13% for models. These results suggest that incorporating contrastive learning techniques into the generation
process of code representation as a preprocessing step can enhance performance in code analysis
RG‑ACA: Efficient and Adaptive Routing Method for Internet of Things Based on Metaheuristic Approach
No full textIn Internet of Things (IoT) systems that use various
and limited devices, efficient use of resources is critical.
However, since this problem is inherently complex, an
enhanced meta-heuristic approach is proposed. In this paper,
a new method called Reverse Gauss Ant Colony Algorithm
(RG-ACA) is suggested for the design of efficient routing
protocol for these systems. It is analyzed on the network
lifetime, throughput and packet delivery parameters and the
results are compared with HEEL, I-HEEL, EES-LEACH and
iABC algorithms. The RG-ACA algorithm ranked first in all
categories compared to other related current studies with
94%, 96 Kbps and 91% performance in these three parameters,
respectively
Elastic Triangular Plate Dynamics on Unilateral Winkler Foundation: Analysis Using Chebyshev Polynomial Expansion for Forced Vibrations
No full textThis research delves into the intricate dynamic and static characteristics of an
elastic triangular plate supported by a unilateral Winkler foundation, with a specific focus on
forced vibrations. The study considers the triangular plate under both uniformly distributed load
and eccentrically applied concentrated load scenarios. The governing equation governing the
plate's behavior is established through a meticulous analysis of its static and dynamic responses.
Utilizing a series of Chebyshev polynomials to represent admissible displacement
functions and employing Lagrange equations of motion, we derive a comprehensive understanding
of the system's dynamics. Due to the non-linear nature of the unilateral Winkler foundation,
an iterative numerical solution methodology is devised. Detailed numerical investigations
shed light on the static behavior of the plate under concentrated loads, exploring a broad
spectrum of parameters encompassing plate geometry and foundation stiffness. Transitioning
to dynamic analyses in the time domain, we adopt a stepwise time variation approach for concentrated
loads, employing constant acceleration procedures to solve the governing differential
equations. Through visual representations, we offer insights into the time variations of contact
regions and plate displacements across various foundation and plate parameters, emphasizing
the non-linear effects arising from plate lift-off phenomena. Extensive exploration of parameter
and loading effects underscores the profound influence of unilateral foundation properties on
both static and dynamic triangular plate behaviors
Resveratrol-Loaded PCL-PEG/GO/HAP Biocomposite Bone Membranes: Evaluation of Mechanical Properties, Release Kinetics and Cellular Response
Get PDFIn this study, biocomposite membranes were developed by incorporating resveratrol (RSV)-loaded PCL-PEG composites, modified with graphene oxide (GO) and hydroxyapatite (HAP). The aim was to enhance hydrophilicity with GO and improve bioactivity with HAP. The release kinetics of RSV was evaluated by using Franz diffusion cells and compared with various kinetic models, including Korsmeyer-Peppas, Higuchi, and Baker, all of which showed high correlation coefficients (R²) close to 0.99. Mechanical tests was performed to determine the suitability of these membranes for tissue engineering applications. The composite membrane modified with GO and HAP exhibited tensile strength of 105.2 ± 5.8 MPa, tensile modulus of 3895 ± 159 MPa, elongation at break of 8.4 ± 0.9%, and toughness of 5.88 ± 0.46 MJ/m³. In vitro cell adhesion studies, visualized using DAPI fluorescence staining, demonstrated increased cell adhesion to the composite membranes over periods of 1, 3, 5, 7, and 14 days. These findings highlight the potential of the RSV-loaded PCL-PEG membranes, enhanced with GO and HAP, for applications in bone tissue engineering