Deep learning models achieve 98.91% accuracy in emotion recognition from physiological signals

Researchers have developed a deep learning approach for recognizing emotions from physiological signals, achieving a high accuracy of 98.91%. The study evaluated Long Short-Term Memory (LSTM), Temporal Convolutional Networks (TCN), and Transformer models using the WESAD dataset, which includes data from wrist and chest sensors. Findings indicate that Transformer models perform best with multimodal inputs, while TCNs excel with wrist-only data. An ensemble method combining predictions from all architectures and modalities yielded the highest overall performance. AI

IMPACT Demonstrates advanced deep learning techniques for physiological emotion recognition, potentially improving health monitoring and affective computing systems.

RANK_REASON Research paper published on arXiv detailing novel deep learning models for emotion recognition. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• Desta Haileselassie Hagos
• Hugging Face
• long short-term memory
• Temporal Convolutional Networks
• transformer
• WESAD

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →