[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-35079":3,"doc-seo-35079":29},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},35079,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","Federated Meta-Learning and Explainable AI for Hybrid Multimodal Emotion Recognition","Emotion recognition supports healthcare, human–computer interaction, and smart environments, but current approaches are constrained by limited data across multiple sensors, poor generalization across datasets, and a lack of interpretability that reduces user trust. This research framework integrates Federated Meta-Learning (FML) with Explainable AI (XAI) to improve accuracy, adaptability, and transparency. It fuses EEG, ECG, GSR, speech, and facial images, extracts features via VMD, wavelet transforms, and deep learning, and enables cross-dataset learning without sharing private data. Explanations leverage SHAP and LIME to clarify model decisions. The approach targets real-time health monitoring and driver-fatigue applications with dependable scalability.","","cbCaidB5elcTecn1","https://ap.wps.com/l/cbCaidB5elcTecn1","pdf",1365474,1,14,"English","en",105,"# Introduction\n# Methodology\n## Federated Meta-Learning\n## Hybrid Multimodal Feature Extraction\n## Explainable AI with SHAP and LIME\n# Application Scenarios\n## Real-Time Health Monitoring\n## Driver Fatigue","[{\"question\":\"What problem does federated meta-learning address in multimodal emotion recognition?\",\"answer\":\"It enables models to learn from separate datasets and share results without exchanging private data, improving adaptability across data sources.\"},{\"question\":\"Which data modalities are used in the proposed hybrid emotion recognition framework?\",\"answer\":\"The framework combines EEG, ECG, GSR, speech, and facial images to capture complementary physiological and behavioral cues.\"},{\"question\":\"How does the research provide explanations for AI emotion predictions?\",\"answer\":\"It applies SHAP and LIME to generate interpretable insights, helping users understand and trust the model’s decisions.\"}]",1782509077,35,null]