5916 research outputs found
Sort by
Dyadic examination of parents' general psychological distress and coparenting in families with young children: The mediating role of couple satisfaction
No full textThe current research explored the dyadic relationships between general psychological distress (GPD) and coparenting dimensions (cooperation, conflict, triangulation) through the mediation of couple satisfaction among parents with young children. The sample comprised 184 heterosexual couples (184 mothers, 184 fathers, age range from 25 to 57 years) married for 10 years on average. The actor-partner interdependence model (APIM) and APIM Mediation Model analyses demonstrated significant relationships between mothers' and fathers' GPD and all three of their own coparenting dimensions (direct actor effects), also through their own couple satisfaction (indirect actor-actor effects). Additionally, mothers' GPD had direct effects on fathers' coparenting cooperation (partner effect). Fathers' GPD had significant indirect effects on all dimensions of mothers' coparenting through mothers' couple satisfaction (partner-actor effects), plus on mothers' coparenting triangulation through fathers' couple satisfaction (actor-partner effect). Findings were in line with Family System Theory and consistent with prior research. Clinical implications were discussed.Publisher versio
Enhancing service quality and accessibility in airports: Insights from automated social media analysis
No full textAnalyzing user-generated content from social media can provide insight for organizations into how customers evaluate services and identify potential pitfalls. This paper explores approaches for analyzing content published by users on X with a main goal of finding messages mentioning airport services to use as feedback for improving airport service quality (ASQ). A dataset of over 1 million tweets mentioning international airports was cleaned, explored, and preprocessed for the task of detecting topics related to airport services, using a rule-based technique, and off-the-shelf classification algorithms. The topics are based on the Airport Council International (ACI) ASQ measurement framework. The authors studied the presence of various airport service categories in the messages, paying special attention to accessibility. Thereafter, sentiment analysis was performed to assess customer perception. Finally, five machine-learning algorithms for multi-label classification were experimented with to detect airport services. The study shows that more than 65% of messages in collected data did not mention any airport services. Around 0.6% of messages concern accessibility, primarily focusing on the process of reaching the airport. Three tested models achieve accuracy, comparable to keyword-based labeling. Additionally, a method for evaluating and comparing sentiment across various airport service categories was proposed, based on automatic sentiment analysis scores
Spectral radiation modeling of gas and soot in glass melting furnaces: Evaluation of RC-SLW vs. WSGG model
No full textAccurate modeling of thermal radiation is crucial for optimizing heat transfer in glass melting furnaces, as it significantly influences flame behavior within the combustion space. Despite its importance, existing literature lacks comparative studies that solve radiation with spectral accuracy and examine different spectral radiation models in combustion atmospheres. This study addresses this gap by comparing the Rank-Correlated Spectral Line-based Weighted Sum of Gray Gases (RC-SLW) model with gray and non-gray WSGG models. The objective is to investigate the impact of different radiation models on combustion behavior and radiative heat transfer in an industrial-scale glass furnace. The combustion process is modeled using a partially premixed steady diffusion flamelet model, the Moss-Brookes soot mechanism for soot formation and oxidation, and the k-ω SST turbulence model. Spectral radiation effects are assessed by implementing the RC-SLW model in ANSYS Fluent through a user-defined function. The RC-SLW model is validated with a 1-D simulation and compared against the line-by-line model solution. Ten cases are examined. The results show that the gray WSGG model overestimates incident radiation by 10% compared to the RC-SLW model, with this discrepancy increasing to 25% when soot radiation is considered. The gray WSGG model also underestimates crown temperatures by up to 11%, while RC-SLW predictions closely align with experimental data, showing a maximum deviation of 6%. Non-gray WSGG models predict similar incident radiation values as the soot-inclusive RC-SLW model (approximately 40% of total energy), but crown temperatures differ significantly. Soot-only WSGG models show temperature discrepancies of up to 20%. These findings underline the importance of accurate spectral radiation modeling to improve furnace simulation accuracy and optimize heat transfer efficiency. © 2025TÜBİTA
Engaging the next generation of Child–Computer Interaction researchers: Teaching CCI
No full textChild–computer Interaction (CCI) is an established research field with its own conference and journal for disseminating results and community engagement. However, there are voices calling for the field to mature (e.g. Torgersson, Bekker, Barendregt, Eriksson, and Frauenberger (2019)). One way to address that is to discuss the ever evolving CCI curriculum in order to discover what efforts are made for engaging a new generation of researchers in the field. This special issue seeks to gather informative and inspirational perspectives on teaching CCI in order to engage the next generation of researchers in the field. Our hope is that this special issue will not only serve as knowledge sharing within the community, but also inspire more people to start teaching CCI and thereby open up for engaging the next generation of researchers
Exploring the nexus of technology and food practices in young adults: A value-sensitive design perspective towards human-food interaction
No full textThis study aims to investigate the dynamics of young adults' food practices within the interplay between technology and tradition by employing a Value-Sensitive Design (VSD) approach as an analytical lens. We conducted surveys and interviews with young adults. This is complemented by a workshop involving design and gastronomy experts. Grounded in Value-Sensitive Design (VSD) and encompassing 38 core values and resulting in five value conflicts. Our analysis highlights five prominent themes: "Preservation of Culinary Heritage and Relationships", "Technological Convenience", "Uniqueness and Personalisation", "Globalised Nature of Food", and "Sustainable Choices and Trustworthiness". By bridging between Human-Food Interaction (HFI) and VSD realms, this study provides insights for researchers, designers, and practitioners. The value-laden analytical perspective of the findings sheds light on the food practices of the young generation, for future HFI design studies blending traditional and technological elements.Publisher versio
Federated matrix factorization for media recommender systems medya neri sistemleri i in federe matris fakt rizasyonu
No full textWith the widespread adoption of technology, the importance of data privacy is increasing, requiring a reassessment of machine learning models. One proposed solution to protect personalized user experiences is Federated Learning (FL). FL eliminates the need for centralized storage by keeping data on the devices where it is generated and ensures privacy through a distributed learning approach. Within an FL framework, user data is stored on the device, and the model learns general patterns across all clients without accessing the actual content. In this study, Matrix Factorization (MF), a Collaborative Filtering method used in recommender systems, was implemented within the Flower FL framework in a decentralized setting. MF was trained using Stochastic Gradient Descent (SGD) and simulated with the Movielens dataset. This study shows that privacy-preserving recommender systems can be developed using FL, and performance can be improved by tuning FL parameters and aggregation strategies
Pagerank-based unsupervised deep vertex representations for anti-money laundering detection
No full textAnti-money laundering (AML) refers to a global framework of laws, regulations, and protocols designed to trace and expose illicit funds that have been concealed to appear legitimate. Enforcing anti-money laundering measures demands rigorous monitoring of financial transactions to detect potentially unlawful activities. However, conventional rule-based AML systems, while compliant with regulatory standards, generate excessive false positives that burden analysts and obscure subtle laundering patterns. To address these limitations, advanced machine learning techniques, particularly those based on graph structures like Graph Neural Networks (GNNs), offer a promising alternative. In this context, we introduce a novel method, referred to as AML PD (Anti-Money Laundering with Personalized Diffusion), which utilizes Personalized PageRank and diffusion mechanisms to generate unsupervised node embeddings. This technique learns inductive graph embeddings for AML detection by taking into account the directionality of graph edges and incorporating both node and edge features, along with the graph's structural information. AML PD derives a node's local representation by fusing diffusion processes with Personalized PageRank scores. This fusion captures critical information relevant to AML when transformed into a lower-dimensional embedding space. These embeddings are then used by a classifier to identify potential instances of money laundering. Our method is not only scalable to large datasets but also enables deeper analysis of the learned embeddings. Experimental results indicate that our method surpasses existing baseline models, particularly when positional encodings are integrated. A core objective of our work is to enhance the efficiency of AML investigations by offering AI-powered insights. As such, AML PD demonstrates strong potential in improving the performance of GNN-based AML detection systems.Publisher versio
Strategic analysis of digital service quality in digital transformation of healthcare using intuitionistic fuzzy AHP
No full textIn recent years, all sectors have been undergoing dramatic transformation. One of the areas that has accelerated this transformation is the healthcare sector. First, the health landscape of the world began to change with the pandemic. This paved the way for digital technologies such as Internet of Things (IoT), artificial intelligence (AI), augmented reality (AR), virtual reality (VR), and blockchain to take place in the healthcare sector and accelerated digital transformation (DT). These digital technologies have changed the way healthcare services are delivered, provided, and managed. DT in healthcare offers advantages such as security, data use and management, operational efficiency, and low costs, while enabling patients and customers to benefit from the opportunities provided by digital technologies. E-health, telehealth, telemedicine, virtual health, and wearable innovative technologies have started to take their place among the new trends in the healthcare sector. Thus, expectations and perceptions regarding health ser-vices have completely changed. Measuring the service quality (SQ) provided by health service providers has moved away from traditional approaches. From this point of view, this study aims to strategically analyze digital service quality (DSQ) in digital healthcare. In this regard, this study conducts an application in Turkey using the intuitionistic fuzzy analytic hierarchy process (IF AHP) method to calculate and prioritize the weights of the criteria affecting DSQ in healthcare.Özyeğin Üniversites
A column generation-based approach for the adaptive stochastic blood donation tailoring problem
No full textManaging blood donations is a challenging problem due to the perishability of blood, limited donor pool, deferral time restrictions, and demand uncertainty. The problem addressed here combines two important aspects of blood supply chain management: the inventory control of blood products and the donation schedule. We propose a stochastic scenario-based reformulation of the blood donation management problem that adopts multicomponent apheresis and utilises donor pool segmentation into here-and-now and wait-and-see donors. We propose a flexible donation scheme that is resilient against demand uncertainty. This scheme enables more flexible donation schedules because wait-and-see donors may adjust their donation schedules according to the realised values of demand over time. We propose a column generation-based approach to solve the associated multi-stage stochastic donation tailoring problem. The numerical results show the effectiveness of the proposed optimisation model, which provides solutions with less than a 7% optimality gap on average with respect to a lower bound. It also improves the operational cost of the standard donation scheme that does not use wait-and-see donors by more than 18% on average. Utilising multicomponent apheresis and flexible wait-and-see donations are suggested for donation organisations because they yield significant cost reductions and resilient donation schedules
Forced migration and the politics of belonging: Integration policy, national debates and migrant strategies
No full textThis research note examines the politics of refugee belonging in Germany, Sweden, Austria, United Kingdom, Italy, Greece, and Turkey. Specifically, it explores how migrant belonging is impacted by integration policies and national political debates on immigration in these countries. Prior research suggests that refugees have little knowledge of policy, but that national political or media debates strongly impact a feeling of inclusion. Our research shows that both policy and national/media debates affect belonging. Despite widely differing legal and national contexts, the countries studied largely base integration on principles of cultural assimilation that can be hostile to “outsiders” and lead to insecure and contradictory belonging. The article also examines the strategies migrants adopt to forge belonging, depending on the national context. We find that in some contexts, migrants emphasize that they take individual responsibility for integrating and in others they build belonging on cultural and religious similarities with the host community. Thus, this research shows that the national policy environment not only impacts belonging, but also shapes the strategies migrants adopt to achieve it. The research is based on a long-term study conducted as a part of an EU Horizon 2020 project, RESPOND.European Commissio