{"id":"W4382318396","doi":"10.1609/aaai.v37i12.26698","title":"Everyone’s Voice Matters: Quantifying Annotation Disagreement Using Demographic Information","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Annotation; Computer science; Demographics; Variance (accounting); Perspective (graphical); Task (project management); Artificial intelligence; Natural language processing; Politeness; Process (computing); Linguistics; Sociology","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":[],"consensus_categories":[],"category_scores_codex":[0.0006591349,0.0001796877,0.0001761932,0.0003456964,0.0002609115,0.0003766329,0.001315788,0.00006919441,0.0000132597],"category_scores_gemma":[0.0002050637,0.0001483436,0.00009631286,0.001477251,0.00009872359,0.001714132,0.000393988,0.0001979923,0.0001789173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005414269,"about_ca_system_score_gemma":0.00006158248,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008980414,"about_ca_topic_score_gemma":0.000009181028,"domain_scores_codex":[0.9980668,0.00001752865,0.0006264972,0.0003238904,0.0006253381,0.0003399412],"domain_scores_gemma":[0.998668,0.00006713926,0.0004309677,0.0002842775,0.0004890088,0.00006062453],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004323441,0.00008217127,0.002981343,0.0002256869,0.00003518,6.495215e-7,0.007645694,0.007000739,0.06388892,0.8060293,0.0003995365,0.1116676],"study_design_scores_gemma":[0.00002617575,0.00004863428,0.00124219,0.000276591,0.00001096239,0.000002781721,0.001142239,0.8856031,0.06996884,0.04138257,0.00009154108,0.0002044063],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5717425,0.00001081087,0.4211082,0.00491028,0.0008208004,0.0004905803,0.000004083,0.0002664023,0.0006462757],"genre_scores_gemma":[0.9932501,0.00001976625,0.006200557,0.0004318994,0.00004305872,0.00001665781,0.000001383971,0.000007465432,0.00002908586],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8786023,"threshold_uncertainty_score":0.6049278,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1677698905391123,"score_gpt":0.3288002364177132,"score_spread":0.1610303458786008,"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."}}