Lisansüstü Eğitim Enstitüsü Diğer Yayınlar Koleksiyonu

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/11729/6359

Listeleniyor 1 - 20 / 50

Adaptive locally connected recurrent unit (ALCRU)
(Springer Science and Business Media Deutschland GmbH, 2025-07-03) Özçelik, Şuayb Talha; Tek, Faik Boray
Research has shown that adaptive locally connected neurons outperform their fully connected (dense) counterparts, motivating this study on the development of the Adaptive Locally Connected Recurrent Unit (ALCRU). ALCRU modifies the Simple Recurrent Neuron Model (SimpleRNN) by incorporating spatial coordinate spaces for input and hidden state vectors, facilitating the learning of parametric local receptive fields. These modifications add four trainable parameters per neuron, resulting in a minor increase in computational complexity. ALCRU is implemented using standard frameworks and trained with back-propagation-based optimizers. We evaluate the performance of ALCRU using diverse benchmark datasets, including IMDb for sentiment analysis, AdditionRNN for sequence modelling, and the Weather dataset for time-series forecasting. Results show that ALCRU achieves accuracy and loss metrics comparable to GRU and LSTM while consistently outperforming SimpleRNN. In particular, experiments with longer sequence lengths on AdditionRNN and increased input dimensions on IMDb highlight ALCRU’s superior scalability and efficiency in processing complex data sequences. In terms of computational efficiency, ALCRU demonstrates a considerable speed advantage over gated models like LSTM and GRU, though it is slower than SimpleRNN. These findings suggest that adaptive local connectivity enhances both the accuracy and efficiency of recurrent neural networks, offering a promising alternative to standard architectures.
ANN activation function estimators for homomorphic encrypted inference
(Institute of Electrical and Electronics Engineers Inc., 2025-06-13) Harb, Mhd Raja Abou; Çeliktaş, Barış
Homomorphic Encryption (HE) enables secure computations on encrypted data, facilitating machine learning inference in sensitive environments such as healthcare and finance. However, efficiently handling non-linear activation functions, specifically Sigmoid and Tanh, remains a significant computational challenge for encrypted inference using Artificial Neural Networks (ANNs). This study introduces a lightweight, ANN-based estimator designed to accurately approximate activation functions under homomorphic encryption. Unlike traditional polynomial and piecewise linear approximations, the proposed ANN estimators achieve superior accuracy with lower computational overhead associated with bootstrapping or high-degree polynomial techniques. These estimators are trained on plaintext data and seamlessly integrated into encrypted inference pipelines, significantly outperforming conventional methods. Experimental evaluations demonstrate notable improvements, with ANN estimators enhancing accuracy by approximately 2% for Sigmoid and up to 73% for Tanh functions, improving F1-scores by approximately 2% for Sigmoid and up to 88% for Tanh, and markedly reducing Mean Square Error (MSE) by up to 96% compared to polynomial approximations. The ANN estimator achieves an accuracy of 97.70% and an AUC of 0.9997 when integrated into a CNN architecture on the MNIST dataset, and an accuracy of 85.25% with an AUC of 0.9459 on the UCI Heart Disease dataset during ciphertext inference. These results underscore the estimator’s practical effectiveness and computational feasibility, making it suitable for secure and efficient ANN inference in encrypted environments.
Intelligent health monitoring in 6G networks: machine learning-enhanced VLC-based medical body sensor networks
(Multidisciplinary Digital Publishing Institute (MDPI), 2025-05-23) Antaki, Bilal; Dalloul, Ahmed Hany; Miramirkhani, Farshad
Recent advances in Artificial Intelligence (AI)-driven wireless communication are driving the adoption of Sixth Generation (6G) technologies in crucial environments such as hospitals. Visible Light Communication (VLC) leverages existing lighting infrastructure to deliver high data rates while mitigating electromagnetic interference (EMI); however, patient movement induces fluctuating signal strength and dynamic channel conditions. In this paper, we present a novel integration of site-specific ray tracing and machine learning (ML) for VLC-enabled Medical Body Sensor Networks (MBSNs) channel modeling in distinct hospital settings. First, we introduce a Q-learning-based adaptive modulation scheme that meets target symbol error rates (SERs) in real time without prior environmental information. Second, we develop a Long Short-Term Memory (LSTM)-based estimator for path loss and Root Mean Square (RMS) delay spread under dynamic hospital conditions. To our knowledge, this is the first study combining ray-traced channel impulse response modeling (CIR) with ML techniques in hospital scenarios. The simulation results demonstrate that the Q-learning method consistently achieves SERs with a spectral efficiency (SE) lower than optimal near the threshold. Furthermore, LSTM estimation shows that D1 has the highest Root Mean Square Error (RMSE) for path loss (1.6797 dB) and RMS delay spread (1.0567 ns) in the Intensive Care Unit (ICU) ward, whereas D3 exhibits the highest RMSE for path loss (1.0652 dB) and RMS delay spread (0.7657 ns) in the Family-Type Patient Rooms (FTPRs) scenario, demonstrating high estimation accuracy under realistic conditions.
Retinal disease diagnosis in OCT scans using a foundational model
(Springer Science and Business Media Deutschland GmbH, 2025) Nazlı, Muhammet Serdar; Turkan, Yasemin; Tek, Faik Boray; Toslak, Devrim; Bulut, Mehmet; Arpacı, Fatih; Öcal, Mevlüt Celal
This study examines the feasibility and performance of using single OCT slices from the OCTA-500 dataset to classify DR (Diabetic Retinopathy) and AMD (Age-Related Macular Degeneration) with a pre-trained transformer-based model (RETFound). The experiments revealed the effective adaptation capability of the pretrained model to the retinal disease classification problem. We further explored the impact of using different slices from the OCT volume, assessing the sensitivity of the results to the choice of a single slice (e.g., “middle slice”) and whether analyzing both horizontal and vertical cross-sectional slices could improve outcomes. However, deep neural networks are complex systems that do not indicate directly whether they have learned and generalized the disease appearance as human experts do. The original dataset lacked disease localization annotations. Therefore, we collected new disease classification and localization annotations from independent experts for a subset of OCTA-500 images. We compared RETFound’s explainability-based localization outputs with these newly collected annotations and found that the region attributions aligned well with the expert annotations. Additionally, we assessed the agreement and variability between experts and RETFound in classifying disease conditions. The Kappa values, ranging from 0.35 to 0.69, indicated moderate agreement among experts and between the experts and the model. The transformer-based RETFound model using single or multiple OCT slices, is an efficient approach to diagnosing AMD and DR.
Relationships among organizational-level maturities in artificial intelligence, cybersecurity, and digital transformation: a survey-based analysis
(Institute of Electrical and Electronics Engineers Inc., 2025-05-19) Kubilay, Burak; Çeliktaş, Barış
The rapid development of digital technology across industries has highlighted the growing need for enhanced competencies in Artificial Intelligence (AI), Cyber security (CS), and Digital Transformation (DT). While there is extensive research on each of these domains in isolation, few studies have investigated their relationship and joint impact on organizational maturity. This study aims to address this gap by analyzing the relationships among the maturity levels of AI, CS, and DT at the organizational level using Structural Equation Modeling (SEM) and descriptive statistical methods. A mixed-methods design combines quantitative survey data with synthetic modeling techniques to assess organizational preparedness. The findings demonstrate significant bidirectional correlations among AI, CS, and DT, with technology and finance being more advanced than government and education. The research highlights the necessity of an integrated AI-CS strategy and provides actionable recommendations to increase investments in these domains. In contrast to the preceding fragmented evaluations, the current research establishes a comprehensive, empirically grounded framework that acts as a strategic reference point for digital resilience. Follow-up studies will involve collecting real-world industry data in support of empirical validation and predictive ability in measuring AI and CS maturity. This research adds to the existing literature by filling the gaps among fragmented digital maturity models and providing a consistent empirical base for organizations to thrive in an evolving technological environment.
Grammar or crammer? the role of morphology in distinguishing orthographically similar but semantically unrelated words
(Institute of Electrical and Electronics Engineers Inc., 2025) Ercan, Gökhan; Yıldız, Olcay Taner
We show that n-gram-based distributional models fail to distinguish unrelated words due to the noise in semantic spaces. This issue remains hidden in conventional benchmarks but becomes more pronounced when orthographic similarity is high. To highlight this problem, we introduce OSimUnr, a dataset of nearly one million English and Turkish word-pairs that are orthographically similar but semantically unrelated (e.g., grammar - crammer). These pairs are generated through a graph-based WordNet approach and morphological resources. We define two evaluation tasks - unrelatedness identification and relatedness classification - to test semantic models. Our experiments reveal that FastText, with default n-gram segmentation, performs poorly (below 5% accuracy) in identifying unrelated words. However, morphological segmentation overcomes this issue, boosting accuracy to 68% (English) and 71% (Turkish) without compromising performance on standard benchmarks (RareWords, MTurk771, MEN, AnlamVer). Furthermore, our results suggest that even state-of-the-art LLMs, including Llama 3.3 and GPT-4o-mini, may exhibit noise in their semantic spaces, particularly in highly synthetic languages such as Turkish. To ensure dataset quality, we leverage WordNet, MorphoLex, and NLTK, covering fully derivational morphology supporting atomic roots (e.g., '-co_here+ance+y' for 'coherency'), with 405 affixes in Turkish and 467 in English.
Exploring the impact of Flash technique on test anxiety among adolescents
(SAGE Publications Ltd, 2025-07) Çitil Akyol, Canan; İnci İzmir, Sevim Berrin
This study aims to investigate the specific effects of Flash Technique (FT) on adolescents with test anxiety. This follow-up study consists of 38 adolescents, 14–17 years of age (M = 15.39, SD = 1.13). Pre-post assessments were conducted using the Test Anxiety Inventory (TAI), Scale of Attitudes Negatively Affecting the Performance I/Test (POET), and Beck Anxiety Inventory (BAI) at baseline, at the end of the 4thand 12thweeks of therapy. The FT was applied for 12 weeks, with one weekly session as an intervention. As a result of the therapy process, the baseline means of total BAI scores decreased from 25.26 to 2.18; the baseline means of TAI decreased from 149.79 to 39.13, and the baseline mean of POET decreased from 298.47 to 73.84 at the end of the 12th week of therapy. Also, the baseline means of SUD scores decreased from 9.42 to zero at the end of the 12th week of treatment. All the adolescents showed complete improvement after the 12th week of the FT. The study findings showed that the test anxiety symptoms significantly decreased with the treatment of the FT. FT can be an effective intervention for test anxiety in adolescents.
Turkish validity and reliability study of the childhood illness attitude scale
(Routledge, 2025-03) Aktan, Zekeriya Deniz; İnci İzmir, Sevim Berrin; Ünlü, Beyza; Yılmaz Kahraman, İpek Su
Severe forms of health anxiety cause serious dysfunction in people’s lives. Childhood Illness Attitude Scales (CIAS) is an assessment tool used to evaluate childhood health anxiety yet has not been validated for use in Turkey. The study aimed to examine the psychometric properties and factor structure of the Turkish version of the CIAS (CIAS-TR). The scale was administered to 306 children aged between 8 and 15 years. In addition to the CIAS-TR, participants were asked to complete the Screen for Child Anxiety-Related Emotional Disorders (SCARED) and the Pediatric Quality of Life Inventory (PedsQL). To measure test-retest reliability, CIAS-TR was completed by participants 15 days later. Results demonstrated good psychometric properties with high internal consistency and test-retest reliability. A positive correlation with SCARED and a negative correlation with PedsQL. Results from Confirmatory Factor Analysis suggested that a four-factor model best fit the data. The findings of the study indicate that the Turkish adaptation of the CIAS is an appropriate tool for assessing health anxiety in children.
Assessing ChatGPT's accuracy in dyslexia inquiry
(Institute of Electrical and Electronics Engineers Inc., 2024) Eroğlu, Günet; Harb, Mhd Raja Abou
Dyslexia poses challenges in accessing reliable information, crucial for affected individuals and their families. Leveraging chatbot technology offers promise in this regard. This study evaluates the OpenAI Assistant's precision in addressing dyslexia-related inquiries. Three hundred questions commonly posed by parents were categorized and presented to the Assistant. Expert evaluation of responses, graded on accuracy and completeness, yielded consistently high scores (median=5). Descriptive questions scored higher (average=4.9568) than yes/no questions (average=4.8957), indicating potential response challenges. Statistical analysis highlighted the significance of question specificity in response quality. Despite occasional difficulties, the Assistant demonstrated adaptability and reliability in providing accurate dyslexia-related information.
Scenes from the mind of an artist (M Train)
(Bentham Science Publishers, 2023-11-12) Özdem, Gökçe
Patti Smith's M Train resembles a mental train that stops at any station, at any time interval. With ambition and inspiration, Smith takes the reader on a journey between dreams and reality, past and present, books and country. Smith's whole life can be considered a work of art. She is an unconventional artist who reveals herself in her relation to space. In an intertwined experience of time and space, we find Smith reminiscing on life, loss, and pains of creation. Smith's analogy of a clock with no hands refers to a frozen time, a memory where the past and the present coexist. This memory also contains the ties that a person establishes with their physical environment. The subjectivity of experience creates differences in the perception of a space. But how is it possible to resist time in our age of speed? This is what Smith presents to her readers: an infinite present. Smith's memory resists its loss, just as architecture resists time. Architecture witnesses personal and social tragedies and freezes them in time. In this sense, architecture turns into a memory remnant, a trace, and survives by creating a bridge between the past, present, and even the future. Smith's experience of the past in the present also makes it possible to interpret the relationship between architecture and experiential time. In this context, architecture reveals memory space and becomes an important factor in the reproduction of memory. Moreover, it can help revive and maintain memory by constructing new forms of expression. In this regard, personal and social memory emerges as a subject that should be emphasized in architectural research.
TURSpider: a Turkish Text-to-SQL dataset and LLM-based study
(Institute of Electrical and Electronics Engineers Inc., 2024-11-25) Kanburoğlu, Ali Buğra; Tek, Faik Boray
This paper introduces TURSpider, a novel Turkish Text-to-SQL dataset developed through human translation of the widely used Spider dataset, aimed at addressing the current lack of complex, cross-domain SQL datasets for the Turkish language. TURSpider incorporates a wide range of query difficulties, including nested queries, to create a comprehensive benchmark for Turkish Text-to-SQL tasks. The dataset enables cross-language comparison and significantly enhances the training and evaluation of large language models (LLMs) in generating SQL queries from Turkish natural language inputs. We fine-tuned several Turkish-supported LLMs on TURSpider and evaluated their performance in comparison to state-of-the-art models like GPT-3.5 Turbo and GPT-4. Our results show that fine-tuned Turkish LLMs demonstrate competitive performance, with one model even surpassing GPT-based models on execution accuracy. We also apply the Chain-of-Feedback (CoF) methodology to further improve model performance, demonstrating its effectiveness across multiple LLMs. This work provides a valuable resource for Turkish NLP and addresses specific challenges in developing accurate Text-to-SQL models for low-resource languages.
The mediating role of SOC and FSOC on parental stress and sleep quality of parents
(John Wiley & Sons Ltd, 2024-08) Kurukütük, Günsu; Ünver, Buket; Özgür Polat, Pelin
[No abstract available]
Emotion regulation, smoking habits, and addiction among university students in Turkiye
(John Wiley & Sons Ltd, 2024-08) Erkol, Ecem; İçer, Yunus; Çam Çelikel, Feryal; Akçınar, Berna
[No abstract available]
Causal links between patents and economic growth: empirical evidence from OECD countries
(Universidade Nove de Julho-UNINOVE, 2024-08) Özkan Yıldız, Öznur; Görkey, Selda
Objective of the Study: This paper empirically investigates the reciprocal relationship and causality between patents and economic growth. Methodology/Approach: Utilizing the Generalized Method of Moments (GMM) Panel Vector Autoregression (PVAR) and panel VAR-Granger Causality frameworks, the study concentrates on Organisation for Economic Co-operation and Development (OECD) economies where a high fraction of global innovative activities take place. Originality/Relevance: The relationship and causality between patents and economic growth are investigated and evaluated by distinguishing the former variable into patent applications and grants. Main Results: The findings from the GMM panel VAR approach indicate that patent applications and grants significantly affect economic growth, whereas economic activities do not influence patent-related variables. The estimations from the panel VAR-Granger approach confirm these findings by presenting a unidirectional causality from patent applications and grants to economic growth. The impulse-response functions (IRFs) exhibit parallel findings, and further checks validate the stability of the findings obtained. The outcomes of this study point out two crucial implications. First, the impacts of patent applications and grants affect economic growth similarly while the impact of patent grants lasts longer. Second, while patents cause higher economic activity, the latter does not induce innovative activity through patents in the OECD. Theoretical/Methodological Contributions: It would be useful to conduct separate analyses for a selected product, sector, or country by including research and development (R&D) expenditures for different periods, country groups, and analysis methods. Social/Management Contributions: Countries should prioritize the establishment of an effective patent management system that will increase the pace of innovation and the implementation of incentive policies for the development of high-value-added technology products.
Transforming tourism experience: AI-based smart travel platform
(Association for Computing Machinery, 2023) Yöndem, Meltem Turhan; Özçelik, Şuayb Talha; Caetano, Inés; Figueiredo, José; Alves, Patrícia; Marreiros, Goreti; Bahtiyar, Hüseyin; Yüksel, Eda; Perales, Fernando
In this paper, we propose the development of a novel personalized tourism platform incorporating artificial intelligence (AI) and augmented reality (AR) technologies to enhance the smart tourism experience. The platform utilizes various data sources, including travel history, user activity, and personality assessments, combined with machine learning algorithms to generate tailored travel recommendations for individual users. We implemented fundamental requirements for the platform: secure user identification using blockchain technology and provision of personalized services based on user interests and preferences. By addressing these requirements, the platform aims to increase tourist satisfaction and improve the efficiency of the tourism industry. In collaboration with various universities and companies, this multinational project aims to create a versatile platform that can seamlessly integrate new smart tourism units, providing an engaging, educational, and enjoyable experience for users.
Efficient estimation of Sigmoid and Tanh activation functions for homomorphically encrypted data using Artificial Neural Networks
(Institute of Electrical and Electronics Engineers Inc., 2024) Harb, Mhd Raja Abou; Çeliktaş, Barış
This paper presents a novel approach to estimating Sigmoid and Tanh activation functions using Artificial Neural Networks (ANN) optimized for homomorphic encryption. The proposed method is compared against second-degree polynomial and Piecewise Linear approximations, demonstrating a minor loss in accuracy while maintaining computational efficiency. Our results suggest that the ANN-based estimator is a viable alternative for secure machine learning models requiring privacypreserving computation.
Multi-task learning on mental disorder detection, sentiment analysis, and emotion detection using social media posts
(Institute of Electrical and Electronics Engineers Inc., 2024) Armah, Courage; Dehkharghani, Rahim
Mental disorders such as suicidal behavior, bipolar disorder, depressive disorders, and anxiety have been diagnosed among the youth recently. Social media platforms such as Reddit have become popular for anonymous posts. People are far more likely to share on these social media platforms what they really feel like in their real lives when they are anonymous. It is thus helpful to extract people's sentiments and feelings from these platforms in training models for mental disorder detection. This study uses multi-task learning techniques to examine the estimation of behaviors and mental states for early mental disease diagnosis. We propose a multi-task system trained on three related tasks: mental disorder detection as the primary task, emotion analysis, and sentiment analysis as auxiliary tasks. We took the SWMH dataset, which included four main different mental disorders already labeled (bipolar, depression, anxiety, and suicide) and offmychest. We then added labels for emotion and sentiment to the dataset. The observed results are comparable to previous studies in the field and demonstrate that deep learning multi-task frameworks can improve the accuracy of related text classification tasks when compared to training them separately as single-task systems.
Innovation diffusion and technology acceptance theory
(CRC Press, 2024-01-01) Akaiso, Emmanuel
Innovation can be defined as the transformation of ideas, goals, visions, and dreams of invention into valuable goods and services for people to exchange for an amount of money. Something everyone on earth must experience is changing, and many people are too afraid to experience this. As a result, they never move ahead in life because they have allowed themselves to be captives of fear, allowing themselves to be unable to alter their circumstances. After the difficult experience of the past twenty-four months due to COVID-19, the eyes of everyone have been opened to how important change, adaptability, and flexibility are. Looking back, we can see that change is particularly important for everyone. That is exactly where innovation comes in (Stenberg 2016). To paraphrase Steve Jobs, innovation differentiates a leader from a follower.
Human risk assessment of heavy metals present in four motor park soils in Lagos State, Nigeria
(CRC Press, 2024-01-01) Akaiso, Emmanuel
The belief that heavies metal pollution is only gotten high intense industries is a common perception of people living in rural areas (Brown et al. 2003). In reality, nowadays, roadways and automobiles are considered to be one of the largest sources of heavy metals. Lead (Pb), copper (Cu), and zinc (Zn) are the most common heavy metals produced from automobiles. Small or little amounts of other metals like nickel and cadmium are found to originate from carparks, bus stops, the roadside, and automobile exhaust (Brown et al. 2003).
Inequality of income distribution and hopes for democratic consolidation in Nigeria
(CRC Press, 2024-01-01) Bamigboye, Oluwaseyi Mike
Given the experiences of many countries seemingly stuck between authoritarian and democratic systems, momentary success proved transient and failed to achieve consolidation (Tonta 2016). While factors such as legitimacy crisis, the rise of militant sub-nationalist agitations, ethnoreligious and identity conflicts, corruption, institutional failure, electoral crime and violence, insecurity, injustice, and political apathy pose challenges to the formation of political culture and have been heavily researched, the role of income inequality in the processes of democratic consolidation remains under-researched. This begs the question: Does income inequality expressed in weak economic growth, rising inflation, and high unemployment, especially among young people, have far-reaching implications for the survival of Nigeria’s democratic system? Using this West African country as a case study, the author argues that income inequality threatens democratic consolidation and tends to a regression into an authoritarian regime. The choice of Nigeria is informed by doubts which have repeatedly crystallized into debates on the likelihood of consolidation since its democratic transition in 1999.

Güncel Gönderiler