{"id":"W3196374756","doi":"10.1029/2021ea001896","title":"Labeling Poststorm Coastal Imagery for Machine Learning: Measurement of Interrater Agreement","year":2021,"lang":"en","type":"article","venue":"Earth and Space Science","topic":"Tropical and Extratropical Cyclones Research","field":"Earth and Planetary Sciences","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Gulf Research Program; U.S. Geological Survey; Directorate for Geosciences; Office of the Director; Leverhulme Trust","keywords":"Computer science; Inter-rater reliability; Set (abstract data type); Artificial intelligence; Process (computing); Data set; Focus (optics); Training set; Machine learning; Supervised learning; Statistics; Artificial neural network; Mathematics","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":[0.000485099,0.00008436483,0.0001359327,0.00005964811,0.000248502,0.00009358759,0.0001343422,0.0000235015,0.0005402333],"category_scores_gemma":[0.0003183572,0.00006154919,0.00004210446,0.0002693004,0.0003425882,0.0001715251,0.00004292966,0.0001325499,0.00002050143],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000002978616,"about_ca_system_score_gemma":0.0001576874,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008907811,"about_ca_topic_score_gemma":0.008792826,"domain_scores_codex":[0.9985596,0.00003908138,0.0001486274,0.0002862708,0.0006073493,0.0003590876],"domain_scores_gemma":[0.9993248,0.0000579175,0.00003668489,0.0001036452,0.000262076,0.0002148676],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001336211,0.00004347611,0.8324572,0.00008358083,0.00001304666,0.00001786936,0.0002394464,0.0002650023,0.09940002,0.0003243698,0.00006859237,0.06695373],"study_design_scores_gemma":[0.0006292354,0.0007506851,0.9200798,0.00004626261,0.000009922842,0.00002090203,0.0003615492,0.01481004,0.04942789,0.0002937767,0.01336064,0.0002093339],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9942197,0.001421246,0.001320582,0.001266689,0.0001738156,0.0001209294,0.00003219554,0.00001196533,0.001432834],"genre_scores_gemma":[0.9961834,0.00008933199,0.003099953,0.00006084213,0.0000412399,7.940408e-7,0.00001096776,0.000001455346,0.0005120452],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.08762252,"threshold_uncertainty_score":0.5915175,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02925973294962399,"score_gpt":0.2447320153598032,"score_spread":0.2154722824101792,"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."}}