21241 research outputs found
Sort by
ASD-HI: A Parent-Child Interaction Dataset for Automated Assessment of Home Intervention
Get PDFGoogle#Gates Foundation#Hewlett Packard Enterprise#Eedi#VitalSource# duolingo english test#Springer#26th International Conference on Artificial Intelligence in Education -- AIED 2025 -- Palermo -- AX6335679Home-based interventions are vital for supporting young children with autism spectrum disorder (ASD), yet many parents struggle to implement strategies effectively due to limited training. While specialists such as educators and speech-language pathologists provide guidance, real-time feedback outside professional settings remains scarce. To bridge this gap, we leverage advances in AI to support parents through automated assessment. However, training such AI systems requires robust data, which is currently limited. To address this, we created ASD-HI (Autism Spectrum Disorder - Home Intervention), a multi-modal dataset comparing 473 real parent-child interaction videos across three families. ASD-HI supports two core tasks: 1) Strategy Detection, identifying the behavioral strategies parents use, and 2) Fidelity Assessment, assessing the fidelity with which these strategies are implemented. We also propose a prompting-based LLM pipeline as a reference approach. It achieves 74% recall and 50% precision for strategy detection and 60% accuracy for fidelity assessment. Our work lays a foundation for developing AI-driven tools to enhance home interventions and improve outcomes for children with special needs. © 2025 Elsevier B.V., All rights reserved
Urban vs rural mortality due to HIV in the USA using the CDC WONDER database over 22 years: a retrospective study
Get PDFIntroductionHIV remains a persistent epidemic in the USA. Although newer and more effective screening and treatment regimens have reduced mortality, a significant disparity exists in urban and rural areas. This study aims to identify disparities in mortality in urban and rural areas for HIV using the CDC-WONDER database from 1999 to 2020 and to analyze any significant differences in mortality rates as well as demographic variables.MethodologyThe Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiological Research (CDC WONDER) database was used to investigate the Rural and Urban Mortality trends for HIV, ICD-10 codes B20-B24, from 1999-2020. The extracted data was systematically grouped by 10-year age range, gender, census region, and race, and analyzed for trends and patterns over time using the R Studio v4.3.2. package to give a statistically significant result with clear representative figures.ResultsThe study showed that urban areas reported 194,717 deaths from HIV, whereas non-metropolitan/rural areas reported 16,009 deaths. The crude mortality rate (CMR)/100,000 population in urban areas declined from over 0.25 in 1999 to 0.05 in 2020. These rates were however always significantly greater than the CMR in rural areas (0.06 in 1999 and 0.01 in 2020). This disparity was most significant among males, the white race, and those between 35-65 years of age.ConclusionsThis study highlights a statistically significant disparity in urban versus rural mortality rates of HIV especially amongst males, the white race, and people between 35-65 years of age
How Do Quality Culture and Human Capital Influence Innovation Capability in SMEs? Examining the Mediating Role of Knowledge Management Capability
Get PDFSmall and medium-sized enterprises (SMEs) are the main components of emerging economies. Their competitive powers directly influence the economy. One of the competition tools for SMEs is innovation capability. This paper investigates the effects of quality culture and human capital on innovation capability and the mediating role of knowledge management capability on these relationships in SMEs. Data were collected from 227 SMEs in Istanbul via an online survey form and analysed with structural equation modelling. Results revealed that quality culture and human capital positively affect innovation capability, and knowledge management capability mediates these associations
Evaluation of Genetic Diversity of Chestnut (Castanea sativa Mill.) Populations Collected from Different Regions of Türkiye Using SCoT and ISSR Molecular Markers
Get PDFThe sweet chestnut (Castanea sativa Mill.) is a widespread tree species in the Mediterranean region. The aim of this study was to determine the level of genetic diversity in nine sweet chestnut populations naturally distributed in three different biogeographical regions of T & uuml;rkiye using ISSR and SCoT markers. In total, 180 chestnut genotypes were screened with 13 ISSR and 16 SCoT primers giving reproducible and clear bands. The 13 of the ISSR and the 16 of the SCoT primers have shown a total of 178 and 297 amplified bands, respectively. The ISSR markers were found to give higher values in the polymorphism information content (PIC), the resolving power (RP) and the marker index (MI) indices. Mantel test calculated between original/unbiased Nei's genetic distances via ISSR and SCoT markers and the results showed a significant correlation between ISSR and SCoT (p = 0.001), indicating a common genetic heritage among the species. In particular, according to population genetic diversity parameters, it was found that the genetic diversity level of the Duzce population (located in Black Sea region) was higher than other populations. The UPGMA cluster analyses showed that there was no clustering according to biogeographic distribution for either marker method. In conclusion, the results of this study have the potential to contribute to the conservation of chestnut genetic resources in T & uuml;rkiye.DUEBAP [BAP-2021.02.02.1154]This article is extracted from doctorate dissertation entitled Determination Genetic Diversity and Regional Adaptations to Climate Change of Chestnut (Castanea sativa Mill.) Populations in Turkiye, supervised by Assoc. Prof. Dr. & Scedil;emsettin Kulac (PhD Dissertation, Duezce University, Duezce, 2023). The thesis is funded by DUEBAP and the project number is BAP-2021.02.02.1154
The Effect of Mother's Voice, Music Voice and White Noise Methods on Pain and Physical Parameters during Venipuncture in Newborn: A Randomized Controlled Study
Get PDFAim: The study was conducted to determine the effect of recorded mother's voice, music voice and white noise methods during the venipuncture procedure on pain level and physiological parameters in newborns. Material and Methods: The study was a randomized controlled trial. The sample of the study consist of 80 newborns (recorded mother’s voice group=20, music voice group=20, white noise=20, control group=20), according to the result of the power analysis. During the venipuncture process, the newborns in the experimental group were listened to the recorded mother's voice, music voice and white noise, while the newborns in the control group were only given routine venipuncture. Results: When the research results were evaluated, it was determined that the pain levels of the newborns in the recorded mother's voice, music voice and white noise groups were significantly lower during and after the procedure compared to the control group (p<0.05). Pain levels of the recorded mother's voice group were significantly lower than those of the music voice and white noise (p<0.05). When the physiological parameter results were evaluated, it was determined that the most positive result in respiration values were in the white noise group (p<0.05). Conclusion: It was observed that mother's voice, music voice and white noise methods are effective in reducing pain and regulating physiological parameter values during venipuncture in newborns. In line with these results, it is recommended that neonatal healthcare professionals use mother's voice, music sound, and white noise methods as non-pharmacological techniques
Chemical ordering in icosahedral and truncated octahedral CoPt@Ag nanoalloys
Get PDFIn this study, the ternary alloying effects on chemical ordering and structural properties of CoPt@Ag nanoalloys were investigated. The optimization of the chemical ordering was performed for icosahedral (Ih) structures with 309 and 561 atoms, and truncated octahedral (TO) structures with 314 and 586 atoms, as they compete in size. The results show that the most stable compositions of the Ih and TO structures for each size have almost the same Co/Pt ratio. The best chemical ordering structures with smaller cores in the Ih configurations of 309 and 561 atoms generally exhibit irregular and asymmetric cores, while those with larger cores tend to have centered cores. Some compositions present a quasi-Janus arrangement in the icosahedral CoPt@Ag nanoalloys. For the TO compositions of 314 and 586 atoms, when the number of Co atoms exceeds a critical value, structural deformations begin at the surface, and the shell becomes asymmetric. With the further increasing in the number of Co atoms, the symmetries of the cores are broken, leading to deformed structures. Additionally, it was found that Co atoms show a greater preference for mixing with Pt atoms than with Ag atoms
Leveraging Machine Learning Techniques to Predict Cardiovascular Heart Disease
Get PDFCardiovascular diseases (CVDs) remain the leading cause of death globally, underscoring the urgent need for data-driven early diagnostic tools. This study proposes a multilayer artificial neural network (ANN) model for heart disease prediction, developed using a real-world clinical dataset comprising 13,981 patient records. Implemented on the Orange data mining platform, the ANN was trained using backpropagation and validated through 10-fold cross-validation. Dimensionality reduction via principal component analysis (PCA) enhanced computational efficiency, while Shapley additive explanations (SHAP) were used to interpret model outputs. Despite achieving 83.4% accuracy and high specificity, the model exhibited poor sensitivity to disease cases, identifying only 76 of 2233 positive samples, with a Matthews correlation coefficient (MCC) of 0.058. Comparative benchmarks showed that random forest and support vector machines significantly outperformed the ANN in terms of discrimination (AUC up to 91.6%). SHAP analysis revealed serum creatinine, diabetes, and hemoglobin levels to be the dominant predictors. To address the current study's limitations, future work will explore LIME, Grad-CAM, and ensemble techniques like XGBoost to improve interpretability and balance. This research emphasizes the importance of explainability, data representativeness, and robust evaluation in the development of clinically reliable AI tools for heart disease detection
Integration of Sensor and Biometric Data in Shooting Training: An Efficient and Goal-Oriented Approach through an Intelligent Decision Support System
Get PDFAtış eğitimi, yüksek maliyetler, zaman kısıtlamaları ve manuel değerlendirme süreçlerinin sınırlılıkları nedeniyle verimlilik açısından önemli zorluklar sunar. Dahası, kursiyerlerin performansını objektif olarak değerlendirmek genellikle zordur, bu da öğrenme sürecini yavaşlatır. Bu çalışmada, eğitim verimliliğini artırmak ve maliyetleri düşürmek amacıyla hem ateşli silah hem de hedefe entegre edilmiş sensör tabanlı bir sistem geliştirilmiştir. İvmeölçer (ACC) ve jiroskop (GYRO) sensörleri, ateşli silahın dinamik hareketlerini hassas bir şekilde ölçerek geri tepme, titreşim, yön değişiklikleri ve açısal hız gibi kritik verileri gerçek zamanlı olarak yakalar. Ek olarak, sensör donanımlı hedef sistemi her atışın doğruluğunu anında tespit eder ve vuruş veya ıskalar hakkında anında geri bildirim sağlar. Önerilen sistem, sadece ateşli silah hareketlerini izlemekle kalmaz, aynı zamanda daha kapsamlı bir performans analizi sunmak için biyometrik verileri de içerir. Atış performansını doğrudan etkileyen önemli bir biyometrik faktör olan kalp atış hızı, gerçek zamanlı olarak izlenir ve analiz edilir. Bu, eğitmenlerin sadece mekanik hataları değil, aynı zamanda kursiyerlerin psikolojik ve fizyolojik durumlarını da dikkate alarak daha bilinçli ve etkili geri bildirimler sunmalarını sağlar. Ayrıca, toplanan verilerden çıkarılan özelliklerin önemi Random Forest algoritması kullanılarak değerlendirilmiştir. Kalp atış hızının veri kümesindeki varyansın yaklaşık %28'ini oluşturduğu gözlemlenmiştir. Son olarak, Destek Vektör Makineleri (SVM) algoritması kullanılarak atış tahmininde %74'lük bir doğruluk oranına ulaşan bir tahmin modeli geliştirilmiştir.Shooting training presents significant challenges in terms of efficiency due to high costs, time constraints, and the limitations of manual assessment processes. Furthermore, objectively evaluating trainees’ performance is often difficult, which in turn slows down the learning process. In this study, a sensor-based system integrated into both the firearm and the target was developed to enhance training efficiency and reduce costs. Accelerometer (ACC) and gyroscope (GYRO) sensors precisely measure the dynamic movements of the firearm, capturing critical data such as recoil, vibration, directional changes, and angular velocity in real time. Additionally, the sensor-equipped target system instantly detects the accuracy of each shot and provides immediate feedback regarding hits or misses. The proposed system not only monitors firearm movements but also incorporates biometric data to deliver a more comprehensive performance analysis. Heart rate, a key biometric factor that directly influences shooting performance, is monitored and analyzed in real time. This allows instructors to provide more informed and effective feedback by considering not only mechanical errors but also the psychological and physiological states of the trainees. Moreover, the importance of features extracted from the collected data was evaluated using the Random Forest algorithm. It was observed that heart rate accounts for approximately 28% of the variance in the dataset. Finally, a predictive model was developed using the Support Vector Machines (SVM) algorithm, achieving an accuracy rate of 74% in shot prediction
3 Boyutlu Simülasyon Tekniği İle Oluşturulmuş Kent İçi Araç Sirkülasyon Koridoru Örneğinde Mevsimsel Değişimin Görsel Algıya Etkisi
Get PDFPeyzaj mimarlığı meslek disiplininde yapılan görsel algı çalışmaları, doğal ve yapay çevrenin estetik değerini araştırmak, korumak ve geliştirmek için yapılan bir dizi araştırmayı kapsar. Çalışmaların ortak amacı; insanların fiziksel çevreyi nasıl algıladıklarını ve ondan ne şekilde etkilendiklerini anlamaktır. Bu çalışmada; insanların görsel algılarının mevsimler üzerindeki değişimi ve bunlara hangi değişkenlerin etki ettiğini irdelemek amaçlanmıştır. Bu amaç doğrultusunda, araştırmanın daha hızlı yapılmasına imkân verebilecek 3 boyutlu programlardan yararlanılmış ve aynı bakış noktasından 4 mevsimin ayrı ayrı görselleri kurgulanmıştır. Simülasyonların kurgulanmasında Sketchup, Lumion ve Photoshop programlarından yararlanılmıştır. Oluşturulan görseller uzman ve kullanıcı grubunun yer aldığı psikofiziksel yaklaşım modelinden yararlanılarak anket yoluyla değerlendirilmiştir. Microsoft Excel programında bir araya getirilen veriler SPSS programında korelasyon ve aritmetik ortalama analizlerine tabi tutulmuştur. Yapılan analizler sonucunda kişilerin demografik yapıları ile görsel algı parametreleri arasında anlamlı ilişki bulunmuştur. Manzara güzelliği, görsel tercih, eşsiz karakter, bitkisel çeşitlilik, düzen, doğallık, açıklık, renk, uyum, ilginçlik, büyüleyicilik, çekicilik, güven, ulaşılabilirlik parametreleri içinde renk en yüksek algılanan kriter olma özelliğini göstermiş ve bu algı en fazla ilkbahar mevsiminde görülmüştür
Effect of wall cavity ratios on the earthquake behavior of the building in masonry buildings
Get PDFTürkiye'de, şehir merkezlerinde klasik yığma yapıların kullanımı azalmakta, ancak kırsal bölgelerde hâlâ yaygın olarak tercih edilmektedir. 2023 yılı TÜİK verilerine göre Türkiye'deki toplam konut stokunun yaklaşık %15-20'si yığma yapılardan oluşmaktadır. Bu oran kırsal bölgelerde %30'un üzerine çıkabilmektedir. Yığma yapılar, taş, tuğla, kerpiç, briket ve ahşap gibi doğal malzemelerle inşa edilen, kendi ağırlıkları veya harç kullanılarak birleştirilen yapılardır. Tüm yapılarda olduğu gibi Türkiye, aktif fay hatlarına sahip bir ülke olduğu için yığma yapılarda deprem riski altındadır. Bu nedenle, kırsal alanlarda yaygın olan geleneksel yığma yapıların deprem performanslarının değerlendirilmesi depreme dayanıklı yapı tasarımı açısından önem taşımaktadır. Bu çalışmada, depreme maruz kalacak yığma binalarda duvar boşluk oranlarının yığma binaların deprem davranışları üzerindeki etkileri çok yönlü incelenmiştir. Türkiye Bina Deprem Yönetmeliği-2018 (TBDY-2018)'in tasarım şartlarına uygun olarak 100 m² alana sahip referans bir yığma yapı planı hazırlanmıştır. Yığma yapının deprem etkisi altında ki davranışı analiz edilerek, binanın performans durumu, hasar durumu, ötelenme durumu, taban kesme kuvvetleri ve devrilme momentleri gibi parametreler incelenmiştir. Aynı bina planında duvar boşlukları, referans plana göre oransal olarak %0, %10, %20 ve %30 oranlarında artırılarak 5 farklı tek katlı yığma yapı modeli oluşturulmuştur. Bu modellerin TBDY-2018'e göre performans analizleri yapılmış, deprem kuvvetleri, taban kesme kuvvetleri, devrilme momentleri ve ötelenme miktarları gibi parametreler belirlenmiştir. Elde edilen sonuçlar, referans bina modeliyle karşılaştırılmış ve depreme dayanıklılık açısından en uygun model belirlenmeye çalışılmıştır. Sonuç olarak, yığma yapılara etki eden deprem kuvveti açısından %0 duvar boşluğu olan yapının karşıladığı taban kesme kuvveti referans alındığında, %10 duvar boşluklu yapının taban kesme kuvvetine karşı direncinin yaklaşık %9,26 oranında azaldığı, %20 duvar boşluklu yapıda yaklaşık %28,56 oranında azaldığı, %21,4 duvar boşluğu olan referans yapıda yaklaşık %31,52 oranında azaldığı ve %30 duvar boşluğu olan yapıda ise yaklaşık %34,69 oranında azaldığı görülmüştür. Yapılan tüm analiz ve diğer hesap sonuçları tezin ilgili bölümlerinde detaylı olarak verilmiştir. Yığma yapı modellerinin Deprem analizleri STA4-CAD programı kullanılarak yapılmıştır. Tasarım özellikleri ve deprem analiz sonuçları, kırsal bölgelerde yapılması planlanan yığma yapılar için olası modeller olarak sunulmuştur. Bu model sonuçlarına göre tasarlanan yığma yapıların deprem etkisi altında ki davranışlarının nasıl olabileceği konusu ile ilgili fayda sağlayacağı düşünülmektedir.In Turkey, the use of classical masonry structures in city centres is decreasing, but it is still widely preferred in rural areas. According to 2023 TÜİK data, approximately 15-20% of the total housing stock in Turkey consists of masonry structures. This rate can exceed 30% in rural areas. Masonry structures are structures constructed with natural materials such as stone, brick, adobe, briquette and wood, and assembled using their own weight or mortar. As with all structures, Turkey is a country with active fault lines, and therefore, masonry structures are at risk of earthquakes. Therefore, the evaluation of the seismic performance of traditional masonry structures, which are common in rural areas, is important in terms of earthquake-resistant building design. In this study, the effects of wall cavity ratios on the seismic behaviour of masonry buildings that will be exposed to earthquakes were examined in a multifaceted manner. A reference masonry building plan with an area of 100 m² was prepared in accordance with the design conditions of the Turkish Building Earthquake Code-2018 (TBDY-2018). The behaviour of masonry structures under the effect of earthquake was analysed and parameters such as the performance status of the building, damage status, drift status, base shear forces and overturning moments were examined. In the same building plan, 5 different single-storey masonry building models were created by increasing the wall cavity proportionally by 0%, 10%, 20% and 30% compared to the reference plan. Performance analyses of these models were made according to TBDY-2018, and parameters such as earthquake forces, base Shear forces, overturning moments and drift amounts were determined. The obtained results were compared with the reference building model and the most suitable model in terms of earthquake resistance was tried to be determined. As a result, when the base Shear force of the structure with 0% wall cavity is taken as a reference in terms of the earthquake force acting on the masonry structures, it is seen that the resistance of the structure with 10% wall cavity against the base shear force decreases by approximately 9.26%, decreases by approximately 28.56% in the structure with 20% wall cavity , decreases by approximately 31.52% in the reference structure with 21.4% wall cavity and decreases by approximately 34.69% in the structure with 30% wall cavity. All analysis and other calculation results are given in detail in the relevant sections of the thesis. Earthquake analyses of the masonry structure models were performed using the STA4-CAD software program. Design features and earthquake analysis results are presented as possible models for the masonry structures planned to be built in rural areas. It is thought that these model results will be beneficial in terms of how the designed masonry structures will behave under earthquake effects