Browsing by Author "Balli, Tugce"

Now showing 1 - 7 of 7

Citation - WoS: 2
Citation - Scopus: 3
Decoding Functional Brain Data for Emotion Recognition: A Machine Learning Approach
(Assoc Computing Machinery, 2024) Tulay, Emine Elif; Balli, Tugce
The identification of emotions is an open research area and has a potential leading role in the improvement of socio-emotional skills such as empathy, sensitivity, and emotion recognition in humans. The current study aimed at using Event Related Potential (ERP) components (N100, N200, P200, P300, early Late Positive Potential (LPP), middle LPP, and late LPP) of EEG data for the classification of emotional states (positive, negative, neutral). EEG datawere collected from 62 healthy individuals over 18 electrodes. An emotional paradigm with pictures from the International Affective Picture System (IAPS) was used to record the EEG data. A linear Support Vector Machine (C = 0.1) was used to classify emotions, and a forward feature selection approach was used to eliminate irrelevant features. The early LPP component, which was the most discriminative among all ERP components, had the highest classification accuracy (70.16%) for identifying negative and neutral stimuli. The classification of negative versus neutral stimuli had the best accuracy (79.84%) when all ERP components were used as a combined feature set, followed by positive versus negative stimuli (75.00%) and positive versus neutral stimuli (68.55%). Overall, the combined ERP component feature sets outperformed single ERP component feature sets for all stimulus pairings in terms of accuracy. These findings are promising for further research and development of EEG-based emotion recognition systems.
Citation - WoS: 3
Citation - Scopus: 4
Comparative Classification Performances of Filter Model Feature Selection Algorithms in Eeg Based Brain Computer Interface System
(Gazi Univ, Fac Engineering Architecture, 2023) Bulut, Cem; Balli, Tugce; Yetkin, E. Fatih
Brain-computer interface (BCI) systems enable individuals to use a computer or assistive technologies such as a neuroprosthetic arm by translating their brain electrical activity into control commands. In this study, the use of filter-based feature selection methods for design of BCI systems is investigated. EEG recordings obtained from a BCI system designed for the control of a neuroprosthetic device are analyzed. Two feature sets were created; the first set was band power features from six main frequency bands (delta (1.0-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), beta (12-25 Hz), high-beta (25-30Hz) and gamma (30-50 Hz)) and the second set was band power features from ten frequency sub-bands (delta (1-4 Hz), theta (4-8 Hz), alpha1 (8-10 Hz), alpha2 (10-12 Hz), beta1 (12-15 Hz), beta2 (15-18 Hz), beta3 (18-25 Hz), gamma1 (30-35 Hz), gamma2 (35-40 Hz), gamma3 (40-50 Hz)). Ten filter-based feature selection methods are investigated along with linear discriminant analysis, random forests, decision tree and support vector machines algorithms. The results indicate that feature selection methods leads to a higher classification accuracy and eigen value centrality (Ecfs) and infinite feature selection (Inffs) methods have consistently provided higher accuracy rates as compared to rest of the feature selection methods.
Citation - Scopus: 1
Restorative: Improving Accessibility to Cultural Heritage With AI-Assisted Virtual Reality
(Inst Mathematics & Informatics, Bulgarian Acad Sciences, 2025) Balli, Tugce; Peker, Hasan; Piskin, Senol; Yetkin, E. Fatih
Digitalization 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.
Citation - WoS: 2
Citation - Scopus: 1
Neural Signatures of Depression: Classifying Drug-Naive Mdd Patients With Time- and Frequency-Domain Eeg Features During Emotional Processing
(Iop Publishing Ltd, 2025) Sutcubasi, Bernis; Balli, Tugce; Metin, Baris; Tulay, Emine Elif
Accurate classification of major depressive disorder (MDD) remains a significant challenge, particularly because of the confounding effect of medications. This study bridges this gap by focusing on the classification of drug-na & iuml;ve individuals diagnosed with MDD and healthy controls (HCs) using electroencephalogram (EEG) data recorded during emotional processing tasks. This study involved 14 HCs and 14 drug-na & iuml;ve individuals diagnosed with MDD (aged 18-31, 12+ years of education, 12 F/2 M). The participants were presented with positive, neutral, and negative images collected from the International Affective Picture System. The mean power amplitudes of event-related potentials (ERP), including the P200, P300, early, middle, and late components of the late positive potential (LPP), were computed, along with band power features, and used as features for classifiers. A support vector machine model was employed for classification to evaluate the individual contributions of ERP components and band power features and explore the combined effects of ERP components and band power features within themselves. The alpha band power achieved the highest individual classification accuracy among the band power features for negative stimuli (92.86%). The late LPP component was the most discriminative ERP component for positive stimuli, yielding an accuracy rate of 89.29%. Combined analysis of the band power features exhibited high accuracy for both positive and negative stimuli (92.86% each). When the ERP components were combined, the classifier achieved the highest accuracy of 89.29% for both negative and neutral stimuli. Our findings suggest that alpha band power and LPP responses to negative and positive stimuli, respectively, can be used to detect MDD. The comparable performance of individual features to that of the combined feature sets indicates their strength as indicators of emotional processing in MDD. These findings provide valuable insights into the development of more reliable diagnostic tools and treatment monitoring strategies that focus on emotional processing in MDD.
Citation - WoS: 1
Citation - Scopus: 1
Differentiating Functional Connectivity Patterns in Adhd and Autism Among the Young People: a Machine Learning Solution
(Sage Publications inc, 2025) Sutcubasi, Bernis; Balli, Tugce; Roeyers, Herbert; Wiersema, Jan R.; Camkerten, Sami; Ozturk, Ozan Cem; Sonuga-Barke, Edmund
Objective: ADHD and autism are complex and frequently co-occurring neurodevelopmental conditions with shared etiological and pathophysiological elements. In this paper, we attempt to differentiate these conditions among the young people in terms of intrinsic patterns of brain connectivity revealed during resting state using machine learning approaches. We had two key objectives: (a) to determine the extent to which ADHD and autism could be effectively distinguished via machine learning from one another on this basis and (b) to identify the brain networks differentially implicated in the two conditions.Method: Data from two publicly available resting-state functional magnetic resonance imaging (fMRI) resources-Autism Brain Imaging Data Exchange (ABIDE) and the ADHD-200 Consortium-were analyzed. A total of 330 participants (65 females and 265 males; mean age = 11.6 years), comprising equal subgroups of 110 participants each for ADHD, autism, and healthy controls (HC), were selected from the data sets ensuring data quality and the exclusion of comorbidities. We identified region-to-region connectivity values, which were subsequently employed as inputs to the linear discriminant analysis algorithm.Results: Machine learning models provided strong differentiation between connectivity patterns in participants with ADHD and autism-with the highest accuracy of 85%. Predominantly frontoparietal network alterations in connectivity discriminate ADHD individuals from autism and neurotypical group. Networks contributing to discrimination of autistic individuals from neurotypical group were more heterogeneous. These included language, salience, and frontoparietal networks.Conclusion: These results contribute to our understanding of the distinct neural signatures underlying ADHD and autism in terms of intrinsic patterns of brain connectivity. The high level of discriminability between ADHD and autism, highlights the potential role of brain based metrics in supporting differential diagnostics.
Evaluating Cognitive and Emotional Engagement in AI-Assisted Virtual Reality Through EEG
(Inst Mathematics & Informatics, Bulgarian Acad Sciences, 2025) Balli, Tugce; Yetkin, E. Fatih
This study proposes an EEG-based evaluation pipeline for an AI-assisted VR platform designed to deliver immersive cultural heritage experiences for elderly people. EEG data is used to evaluate emotional and cognitive responses while performing real-world versus virtual tasks, offering a reusable evaluation framework for future immersive heritage applications.
Citation - WoS: 1
Citation - Scopus: 1
Verbal Harassment Detection in Online Games Using Machine Learning Methods
(Elsevier Sci Ltd, 2025) Hibatullah, Helmi; Balli, Tugce; Yetkin, E. Fatih
Video games have been an inseparable aspect for many throughout their upbringing. The widespread adoption of the internet in the early 2000s has brought video games from the traditional offline media to the online environment. Consequently, people from different parts of the world can play together and communicate in-game with each other. Nowadays, most massively multiplayer online games (MMOs) incorporate voice communication features. Playing video games online with a certain degree of anonymity, along with the ability to verbally communicate with each other, has proven to be a dangerous combination that can breed toxic and abusive behaviors if left unmoderated. This paper proposes a new approach to integrating Whisper, a pre-trained automatic speech recognition (ASR) model, with the well-researched topic of text-based abusive behavior detection. Our proposed verbal harassment detection pipelines yielded an average F-score of 0.899 for all variants tested.