Decoding Emotional Dynamics: A Comparative Analysis of Contextual and Non-Contextual Models in Sentiment Analysis of Turkish Couple Dialogues

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 172648 - 172695

Published: Jan. 1, 2024

Language: Английский

---

EEG and fNIRS Signal-Based Emotion Identification by Means of Machine Learning Algorithms During Visual Stimuli Exposure

Electronics, Journal Year: 2024, Volume and Issue: 13(23), P. 4797 - 4797

Published: Dec. 5, 2024

This paper presents the identification of arousal and valence during visual stimuli exposure using electroencephalograms (EEGs) functional near-infrared spectroscopy (fNIRS) signals. Specifically, various images were shown to several volunteers evoke different emotions defined by their level valence, such as happiness, sadness, fear, anger. Brain activity was recorded Emotiv EPOC X NIRSport2 devices separately. The signals then processed analyzed identify primary brain regions activated trials. Next, machine learning methods employed classify evoked with highest accuracy values 71.3% for EEG data a Multi-Layer Perceptron (MLP) method 64.0% fNIRS Bagging Trees (BAG) algorithm. approach not only highlights effectiveness technologies but also provides insights into complex interplay between areas emotional experiences. By leveraging these advanced acquisition techniques, this study aims contribute broader field affective neuroscience improve emotion recognition systems. findings could have significant implications developing intelligent systems capable more empathetic interactions humans, enhancing applications in mental health, human–computer interactions, or adaptive environments, among others.

Language: Английский

---

Emerging Neuroimaging Approach of Hybrid EEG-fNIRS Recordings: Data Collection and Analysis Challenges

Data Science in Science, Journal Year: 2024, Volume and Issue: 3(1)

Published: Dec. 6, 2024

The hybrid EEG-fNIRS (electroencephalogram - functional near-infrared spectroscopy) modality provides a comprehensive understanding of brain activity by simultaneously capturing electrical and hemodynamic responses. It takes advantage the temporal resolution in EEG with good spatial fNIRS. This system is non-invasive, portable, relatively affordable compared to magnetic resonance imaging (fMRI). Despite its inherent limitations, systems can be valuable tool neuroimaging, suitable for cognitive clinical research that requires information about function. They are well-suited naturalistic experimental settings, brain-computer interface applications, disease diagnosis. paper also explores electro-metabolic interactions motor cortex during left- right-hand grip tasks. We analyzed cross-frequency coupling (CFC) between delta rhythms (0–4Hz) oxyhemoglobin (HbO) deoxyhemoglobin (HbR) concentrations calculated from CFC was estimated as correlation HbO, HbR, instantaneous amplitude signals. cannot studied non-synchronous measurements not directly related activation amplitudes. Based on previous fMRI-EEG studies, some categories oscillations denote negligible fNIRS However, our results showed contra-lateral delta-EEG HbO significant pattern differences left cortex.

Language: Английский

---

A bidirectional cross-modal transformer representation learning model for EEG-fNIRS multimodal affective BCI

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 266, P. 126081 - 126081

Published: Dec. 10, 2024

Language: Английский

---

CrCo- Mlgcn: A Cross-scale Co-learning Based Multi-Level Graph Convolutional Network for Brain-Computer Interface

Published: Nov. 15, 2024

Language: Английский

---