{"id":"W4320029570","doi":"10.1109/globecom48099.2022.10000909","title":"Annotation Efficiency in Multimodal Emotion Recognition with Deep Learning","year":2022,"lang":"en","type":"article","venue":"GLOBECOM 2022 - 2022 IEEE Global Communications Conference","topic":"Emotion and Mood Recognition","field":"Psychology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Computer science; Wearable computer; Artificial intelligence; Deep learning; Facial expression; Emotion recognition; Emotion classification; Machine learning; Pace; Convolutional neural network; Process (computing); Speech recognition; Human–computer interaction","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008110045,0.0002701013,0.0002728322,0.0002909754,0.001071321,0.00008478178,0.0009503738,0.0001123691,0.005324991],"category_scores_gemma":[0.00008455881,0.0003160172,0.0000904684,0.001747322,0.0001867898,0.0002724347,0.0003642873,0.001096443,0.0003941841],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000614635,"about_ca_system_score_gemma":0.0001435425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001202596,"about_ca_topic_score_gemma":0.002170891,"domain_scores_codex":[0.996015,0.001857569,0.000575125,0.0005971467,0.0004858822,0.0004693343],"domain_scores_gemma":[0.9980465,0.0001598115,0.0003704389,0.001039543,0.0002696577,0.0001140386],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0009599744,0.005120283,0.04536343,0.00005204068,0.0002058525,0.00007143535,0.01286142,0.007834242,0.001132561,0.0246396,0.002460738,0.8992984],"study_design_scores_gemma":[0.02000563,0.005495462,0.4964014,0.0003716147,0.0003844442,0.001334035,0.1418705,0.2542035,0.0002037389,0.0207763,0.05454804,0.004405308],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8509249,0.001145159,0.02356105,0.002186495,0.001645097,0.001612157,0.0005030034,0.0005273818,0.1178947],"genre_scores_gemma":[0.9945139,0.0002399385,0.001668842,0.0003023188,0.00003469818,0.0007544531,0.001975788,0.00002458231,0.0004854471],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8948931,"threshold_uncertainty_score":0.9999292,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05213402463928806,"score_gpt":0.324930110147868,"score_spread":0.2727960855085799,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}