5862 research outputs found
Sort by
Restorative: Improving Accessibility to Cultural Heritage With AI-Assisted Virtual Reality
No full textDigitalization of the cultural heritage can be considered from multiple perspectives. In this work, we present a case study based on the ancient city of Karkemish to propose a structured pipeline for developing an Artificial Intelligence (AI)-assisted Virtual Reality (VR) system. The framework outlines a roadmap for creating a user-friendly and gamified VR interface, incorporating qualitative and quantitative evaluation methods before deployment. Qualitative assessments focus on User Interface/User Experience (UI/UX) design, while quantitative evaluations utilize electroencephalogram (EEG) data to monitor cognitive and emotional responses, aiming to promote a positive user experience. Moreover, we introduce a privacy-preserving approach to ensure the user's privacy during the system interaction. The study's aim is twofold: a) preservation and dissemination of endangered cultural heritages, and b) improving the quality of life for individuals with limited mobility (handicapped, elderly, heritage site restrictions, poverty) by enabling virtual access to cultural heritages
Evaluation of the Selection of Low-Bed Trailers in the Transportation of Oversized and Overweight Cargo: A Hybrid Picture Fuzzy CRITIC-MARCOS Model
No full textProject logistics consists of operational processes that require attention, especially involving the transportation of heavy and sensitive loads. Trailer selection is a critical factor for the successful completion of projects, as trailers play a fundamental role in the safe and efficient transportation of loads. The characteristics of the cargo to be transported and the transportation requirements affect the type and characteristics of the trailer. Choosing the wrong trailer can lead to disruption of projects, additional costs, safety risks, and environmental damage. In addition, the use of improper equipment can compromise road safety and cause legal problems. Therefore, the choice of trailers is of great importance. The presence of many criteria in the selection process complicates this process. The excess of influencing factors requires the use of multi-criteria decision-making methods. This study presents a roadmap to guide decision-makers in selecting trailers for project logistics and heavy transportation using a hybrid decision-making procedure. This procedure combines the Picture Fuzzy CRITIC method, which is a multicriteria decision-making technique that deals with subjective and imprecise information, and the MARCOS method, which is a multi-criteria decisionmaking technique that identifies the best alternative based on the shortest distance from the positive ideal solution and the farthest distance from the negative ideal solution. The results show that carrying capacity is more critical than economic criteria. The results of a comprehensive robustness test confirm the results obtained in the study. © The Author(s) 2025
The Obligation to Inform the Consumer Within the Scope of Legal Design Principles
No full textGünümüzün ekonomik, sosyal ve teknolojik değişimlerinin tüketiciler için yarattığı riskleri ve özellikle satıcı ile tüketici arasındaki 'bilgi asimetrisi' sorunu ile karşılaşılır. Küreselleşme ve dijitalleşme karşısında geleneksel tüketici koruma yöntemleri yetersiz kalmaktadır. Çözüm olarak ise hukuk ile tasarım disiplinlerini birleştiren yenilikçi bir yaklaşım olan 'hukuk tasarımı' bir araç olarak kullanılabilir. Hukuk tasarımı, karmaşık hukuki bilgileri tüketiciler için anlaşılır, erişilebilir ve kullanılabilir hale getirmeyi amaçlar. Bu yaklaşım, insan odaklı tasarım süreçlerini benimseyerek, çok katmanlı bilgilendirme panelleri, görsel simgeler ve etkileşimli arayüzler gibi araçlarla bilginin etkin bir şekilde iletilmesini sağlar. Bilgilendirme yükümlülüğünün sözleşme öncesi, sırası ve sonrasını kapsayan dinamik bir süreç olması gerektiğidir. Amaç, tüketicinin yalnızca şeklen değil, özünde bilgi sahibi olarak rasyonel kararlar vermesini ve sözleşme adaletini sağlamaktır. Özellikle fiziksel temasın olmadığı e-ticaret gibi dijital ortamlarda bu yaklaşımın önemi daha da artmaktadır. Sonuç olarak hukuk tasarımının, tüketici haklarını somut bir şekilde koruyarak, tacir ile tüketici arasında adil bir denge kurarak ve hukuk sistemine olan güveni artırarak çağdaş tüketici hukukunda vazgeçilmez bir araçtır.The economic, social, and technological changes of our time create risks for consumers, particularly encountering the problem of 'information asymmetry' between the seller and the consumer. In the face of globalization and digitalization, traditional consumer protection methods are proving insufficient. As a solution, 'legal design,' an innovative approach that combines the disciplines of law and design, can be used as a tool. Legal design aims to make complex legal information understandable, accessible, and usable for consumers. By adopting human-centered design processes, this approach ensures the effective communication of information through tools such as multi-layered information panels, visual icons, and interactive interfaces. The obligation to inform should be a dynamic process that covers the pre-contractual, contractual, and post-contractual stages. The goal is to ensure that the consumer makes rational decisions not just by being formally informed, but by being substantively knowledgeable, thereby achieving contractual justice. The importance of this approach is even greater in digital environments where there is no physical contact, such as e-commerce. In conclusion, legal design is an indispensable tool in modern consumer law for concretely protecting consumer rights, establishing a fair balance between the merchant and the consumer, and increasing trust in the legal system
Challenges and Opportunities in B2B-to-B2C Shifts: A Comparative Study of Turkish Companies
No full text'B2B'den B2C'ye Geçişlerde Zorluklar Ve Fırsatlar: Türk Şirketleri Üzerine Karşılaştırmalı Bir Çalışma' başlıklı bu tez, bir şirketin B2B (İşletmeden İşletmeye) pazarından B2C (İşletmeden Tüketiciye) pazarına geçişinin mevcut müşteri ağlarının dönüşümü üzerindeki etkisini araştırmayı amaçlamaktadır. Tezin merkezi sorgulaması, bir şirketin yerleşik B2B ağının, B2C modeline yönelik stratejik geçiş sırasında Yeni Ürün Geliştirme (YÜG) ve tasarım yönetimi süreçleri için bir kolaylaştırıcı (avantaj) mı yoksa bir kısıtlayıcı (dezavantaj) olarak mı işlev gördüğünü ve bunun nasıl gerçekleştiğini incelemektedir. Bu çalışma, mevcut ilişkilerin bu ikili doğasını eleştirel bir şekilde inceleyerek, tüketici pazarları için gereken adaptasyonu hem kolaylaştırma hem de engelleme potansiyellerini araştırmaktadır. Çalışma, müşteri ağlarının süreç üzerindeki olumlu veya olumsuz etkilerini ortaya çıkaracak, aynı zamanda şirketlerin karşılaşabileceği zorlukları ve bunların çözümlerini analiz edecektir. Araştırma, işletmelerin bu tür geçiş süreçlerini daha etkin bir şekilde yönetebilmeleri için hem teorik bir çerçeve hem de pratik öneriler sunmayı hedeflemektedir. Bu araştırma, şirketlerin tasarım ve marka yönetimi perspektifinden karşılaştıkları dinamikleri anlamayı amaçlayan nitel bir araştırma olarak yapılandırılmıştır. Bu çalışmada, B2B'den B2C modellerine geçiş yapmış farklı sektörlerden (ambalaj, bisiklet bileşenleri ve ayakkabı) üç Türk şirketini içeren çoklu vaka çalışması yöntemi, zorlukları ve uygulamaları analiz etmek için kullanılmıştır. Ayrıca, araştırma verileri yarı yapılandırılmış mülakatlar yoluyla toplanmıştır. Sonuçlar, şirketin marka genişletme stratejisinin pazar payının büyümesi üzerinde olumlu bir etkisi olduğunu ortaya koymuştur. İşletmenin marka yönetimi, bir pazarlama aracı ve stratejik bir gerekliliktir. Marka yönetiminde tasarımın önemi, markanın vaatleri ile müşteri deneyimi arasında tutarlılık sağlama ihtiyacını vurgulamaktadır.This thesis, titled 'Challenges And Opportunities In B2B-To-B2C Shıfts: A Comparative Study Of Turkish Companies', seeks to explore the effect of a company shifting from B2B (business-to-business) to B2C (business-to-consumer) markets on the transformation of its existing customer networks. The central inquiry investigates whether, and how, a company's established B2B network acts as an enabler (advantage) or a constraint (disadvantage) for its New Product Development (NPD) and design management processes during the strategic shift towards a B2C model. This study critically examines the duality of these existing relationships, exploring their potential to both facilitate and hinder the adaptation required for consumer markets. The study will uncover the positive or negative impacts of customer networks on the process, alongside analyzing the challenges that companies may face and their solutions. The research aims to provide both a theoretical framework and practical recommendations for businesses to manage such transition processes more effectively. This research is structured as qualitative research aiming to understand the dynamics that companies face from the perspectives of design and brand management. In this study, a multi-case study method, involving three Turkish companies from different sectors (packaging, bicycle components, and footwear) that have transitioned from B2B to B2C models, was used to analyze the difficulties and practices. In addition, the research data were collected through semi-structured interviews. The results revealed that the company's brand extension strategy has a positive impact on the growth of its market share. The business' brand management is a marketing tool and a strategic necessity. The importance of design in brand management emphasizes the need to ensure consistency between the brand's promises and the customer experience
Predicting Individual Digital Behavior: From Probabilistic Models to Deep Learning with a Novel Hybrid Framework
No full textToplumun hızla dijitalleşmesi, bireysel davranışları yeniden şekillendirmiş ve dijital etkileşimler aracılığıyla büyük hacimli bireysel düzeyde veriler üretmiştir. Bu dönüşüm, tahmine dayalı modelleme açısından fırsatlar barındırsa da, geleneksel tahmin teknikleri belirli kısıtlamalara sahiptir. Örneğin, Otoregresif Hareketli Ortalama (ARMA) modelleri toplu verilere dayanırken, Ölene Kadar Satın Al (BTYD) modelleri davranışsal eğilimleri modellemekte zorlanır. Bu zorlukların üstesinden gelmek için bu çalışma, bireysel düzeydeki dijital davranışları modellemek amacıyla Geçitli Tekrarlayan Birim (GRU) ve Uzun-Kısa Süreli Bellek (LSTM) modellerini kullanan bir derin öğrenme mimarisi sunmaktadır. LSTM daha önceki çalışmalarda kullanılmış olsa da, GRU bu bağlamda daha önce incelenmemiştir. Bulgularımız, GRU'nun son döneme ait örüntülere duyarlılığı sayesinde, Yakınlık, Sıklık, Parasal Değer (RFM) yaklaşımının yakınlık bileşeniyle uyumlu olarak bireysel dijital davranış tahmininde etkili olduğunu göstermektedir. Finansal teknoloji mobil uygulama verisi ile Online Retail II veri seti üzerinde yapılan ampirik değerlendirmeler, GRU ve LSTM modellerinin davranışsal eğilimleri başarıyla yakaladığını, BTYD yaklaşımlarına kıyasla daha düşük yanlılık ve daha yüksek trend doğruluğu sunduğunu göstermektedir. Bu çalışmada ayrıca, kural tabanlı segmentasyon veya k-ortalama kümeleme yöntemlerini derin öğrenme ve BTYD modelleriyle birleştiren hibrit bir yaklaşım önermektedir. Bu yaklaşım, tek başına kullanılan derin öğrenme modellerine kıyasla belirgin bir performans artışı sağlamasa da, davranışsal modellemede yeni ve incelenmemiş bir yön sunmaktadır. Bu araştırma, metodolojik katkılarının yanı sıra, tahmine dayalı \mbox{modellemede} satın alma dışındaki davranışsal verilerin kullanılmasının önemini de ortaya koymaktadır. Geleneksel tahmin yöntemleri genellikle müşteri satın alma davranış verileri için geliştirilmiş ve uygulanmıştır. Bu çalışma ise kapsamı genişleterek kullanıcı girişleri gibi dijital davranışsal etkileşimleri ele almaktadır. Elde edilen bulgular, dijital davranışların tahmin edilmesinde dizi tabanlı \mbox{modellerin} taşıdığı potansiyele dair teorik içgörüler sunarken, karar alma süreçlerini güçlendirmeyi hedefleyen kuruluşlar için de pratik çıkarımlar ortaya koymaktadır. Bu sonuçlar, doğru hedef kullanıcıları belirleme, kişiselleştirilmiş stratejiler geliştirme, elde tutma kampanyalarını optimize etme ve bireysel davranış değişimlerine proaktif yanıt verme konularında yardımcı olabilir.The rapid digitalization of society has reshaped individual behavior, generating large volumes of individual-level data through digital interactions. While this presents opportunities for predictive modeling, traditional forecasting techniques face limitations. For example, Autoregressive Moving Average (ARMA) models rely on aggregate data, and Buy Till You Die (BTYD) models often struggle to model dynamic behavioral trends. To address these challenges, this study introduces a deep learning framework that leverages Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) architectures to model individual-level digital behavior. While LSTM has been used in prior studies, GRU has not previously been explored in this context. Our findings indicate that GRU is effective in individual-level digital behavior prediction, particularly due to its sensitivity to recent patterns, which aligns with the recency component of the Recency, Frequency and Monetary (RFM) framework. Empirical evaluation using data from a fintech mobile application and the Online Retail II dataset shows that both GRU and LSTM outperform BTYD models in trend fidelity and bias reduction. This study also presents a segmentation-based hybrid approach that combines rule-based segmentation or -means clustering with deep learning and BTYD models. Although the hybrid framework did not produce substantial performance improvements over standalone deep learning models, it introduces an unexplored direction in behavioral modeling. Beyond methodological contributions, this research highlights the importance of incorporating non-purchase behavioral data in predictive modeling. While traditional forecasting methods have primarily been developed and applied for customer purchase behavior data, this study broadens the scope by considering digital behavioral interactions, such as user logins. The findings provide theoretical insight into the potential of sequence-based models in digital behavior prediction and practical implications for organizations aiming to enhance decision-making. Specifically, the results can help organizations identify target users, personalize engagement strategies, optimize retention campaigns and respond proactively to individual-level shifts in behavior
An Experimental Study for Analysing Pricing Decisions Under Competition
No full textAccepted by: M. Zied BabaiAlthough pricing models of revenue management can considerably improve the profitability of firms, it is not evident whether real-life decision-makers follow them precisely in making price decisions. The effect of sequential anchors on the pricing decisions especially remains understudied. This study aims to investigate the decision-making patterns of human beings in setting prices for homogeneous goods in a competitive market. Two laboratory experiments have been designed and conducted, the former involving a competitor firm with a fixed price, while in the latter, the competitor's price changes over time. The results show that decision-makers are more prone to the 'anchoring effect' when they encounter varying competitor prices. This bias could override the learning effect and is not statistically different across the two genders. Moreover, 'underpricing' is frequently observed in this setting, and a higher variance in demand could deteriorate the quality of pricing decisions. The competitor's prices act as sequential anchors, and several insights are derived regarding the size and extent of anchoring tendency under various patterns of sequential anchors. More generally, the results of the experiment bring practical insights regarding how to enhance pricing decisions for managers in stochastic demand settings with varying parameters.TUBITAK (The Scientific and Technological Research Council of Turkey) [218K341]This study received funding from TUBITAK (The Scientific and Technological Research Council of Turkey) in the scope of Project #218K341
Hosting Capacity Calculation Methods
No full textIn this chapter, we focus on hosting capacity (HC) calculations, by giving the methods to determine the maximum amount of distributed energy resources (DER) that can be integrated into power distribution network(s) without compromising reliability or performance. We detail methodologies such as power flow-based approaches, probabilistic techniques, and machine learning algorithms, with sample applications of HC calculations. Initially, we focus on power flow-based methods based on simulating power distribution network(s) to assess system voltage, current flow, and stability impacts from DER installations. Then, we will give the probabilistic approaches that use uncertainties in renewable generation and consumer demand, based on statistical techniques and Monte Carlo simulations aiming to reflect these variability. Machine learning (ML) techniques will also be given based on analyzing large data sets, detecting patterns, and predicting system responses. These kinds of methods include regression analysis and neural networks trained on historical data for optimized HC predictions. It should be stated that HC is impacted by several factors, such as network topology, load profiles, and DER characteristics, and these as well will be discussed. We will provide a practical example of an HC calculation on a 141-node distribution network using a step-by-step algorithm in Matpower, with simulation results based on an iterative deterministic method. Then, we will give the broader implications of HC assessments for grid modernization and energy policy, highlighting how accurate calculations support a more decentralized, sustainable, and resilient energy future. © 2025 Elsevier Inc. All rights reserved
Enhancing Real Estate Listings Through Image Classification and Enhancement: a Comparative Study
No full textWe extended real estate property listings on the online prop-tech platform. On the platform, the images were classified into the specified classes according to quality criteria. The necessary interventions were made by measuring the platform’s appropriateness level and increasing the advertisements’ visual appeal. A dataset of 3000 labeled images was utilized to compare different image classification models, including convolutional neural networks (CNNs), VGG16, residual networks (ResNets), and the LLaVA large language model (LLM). Each model’s performance and benchmark results were measured to identify the most effective method. In addition, the classification pipeline was expanded using image enhancement with contrastive unsupervised representation learning (CURL). This method assessed the impact of improved image quality on classification accuracy and the overall attractiveness of property listings. For each classification model, the performance was evaluated in binary conditions, with and without the application of CURL. The results showed that applying image enhancement with CURL enhances image quality and improves classification performance, particularly in models such as CNN and ResNet. The study results enable a better visual representation of real estate properties, resulting in higher-quality and engaging user listings. They also underscore the importance of combining advanced image processing techniques with classification models to optimize image presentation and categorization in the real estate industry. The extended platform offers information on the role of machine learning models and image enhancement methods in technology for the real estate industry. Also, an alternative solution that can be integrated into intelligent listing systems is proposed in this study to improve user experience and information accuracy. The platform proves that artificial intelligence and machine learning can be integrated for cloud-distributed services, paving the way for future innovations in the real estate sector and intelligent marketplace platforms. © 2025 by the authors.Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, TUBITAK, (7220634); Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, TUBITA
Veterinary Ethics in Practice: Euthanasia Decision Making for Companion and Street Dogs in Istanbul
No full textThis article examines how veterinarians in Istanbul experience and navigate the ethical, emotional, and institutional complexities of performing euthanasia on dogs, with particular attention to the differences between companion and street dogs. Drawing on 29 in-depth interviews with private practice veterinarians in Istanbul, this study employs qualitative analysis using the NVivo 12 Plus software and reflexive thematic analysis to identify key patterns in moral reasoning, emotional labor, and clinical decision making. The findings indicate that euthanasia of companion dogs is often framed through shared decision making with guardians, emotional preparation, and post-procedural grief rituals. While still emotionally taxing, these cases are supported by relational presence and mutual acknowledgment. In contrast, euthanasia of street dogs frequently occurs in the absence of legal ownership, institutional accountability, or consistent caregiving, leaving veterinarians to bear the full moral and emotional weight of the decision. Participants described these cases as ethically distinct, marked by relational solitude, clinical ambiguity, and heightened moral distress. Six key themes that reveal how euthanasia becomes a site of both care and conflict when structural support is lacking are identified in this study, including emotional burden, ethical strain, and resistance to routinized killing. By foregrounding the roles of institutional absence and relational asymmetry in shaping end-of-life decisions, this study contributes to empirical veterinary ethics and calls for more contextually attuned ethical frameworks, particularly in urban settings with large populations of street dogs and culturally entrenched practices of collective guardianship and caregiving.The Scientific and Technological Research Council of Turkey (TUBITAK) [124K093]This research was supported by The Scientific and Technological Research Council of Turkey (TUBITAK) under Grant No. [Project Number: 124K093] within the scope of the 3501-Career Development Program
Bringing Closer the Distant Land Called Elderliness: The Interweaving of Form and Content in Mohamed Al Khatib's La Vie Secrète Des Vieux (The Secret Life of the Elderly)
No full textBu makalede 78. Avignon Festivali'nde yer alan, Mohamed El Khatib'in yazıp yönettiği ve sahnede kendisinin de kolaylaştırıcı olarak yer aldığı La vie secrète des vieux (Yaşlıların Gizli Yaşamı) adlı oyunu incelenmektedir. Yapıt, yaşlıların çocuklaştırılması, huzurevlerinde ve bakımevlerinde aldıkları hizmetlerin niteliği ve sağlık koşullarının aksine sosyolojik inceleme alanları dışında kalan ve yaşlıların yaşamsal sorunları arasında görülmeyen aşk hayatlarını ele alır. La vie secrète des vieux, aşkın insanı yaşama bağlayan nüvesini, aşkın fiziksel yönünü, temasın anlamını huzurevlerinde yaşamakta olan yetmiş beş yaş üstü bireylerin bizzat kendilerinin sahne üstünde bizzat anlattıkları bir çalışmadır. Khatib’in bu oyununun ayırt edici özelliği yaşlıların kendilerini anlatmasına olanak vermesi, toplumun farklı katmanlarının birbirini anlamasına bir zemin oluşturması ve bu yolla kuşaklar arası köprü kurmasıdır. Makalenin amacı belgesel tiyatro ile kamu yararına tiyatro gibi türler arasında sınıflandırılmayı tercih etmeyen Khatib’in tiyatroya yaklaşımını araştırmak, biçim ile içerik ilişkisine odaklanarak sanatçının kendi oluşturduğu özgün biçimin kaynaklarını incelemektir