{"id":"W4410932950","doi":"10.1186/s13750-025-00362-9","title":"Evaluating generative AI for qualitative data extraction in community-based fisheries management literature","year":2025,"lang":"en","type":"article","venue":"Environmental Evidence","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Centre for Marine Socioecology; Canada Excellence Research Chairs, Government of Canada; Commonwealth Scientific and Industrial Research Organisation","keywords":"Context (archaeology); Computer science; Data science; Data extraction; Generative grammar; Systematic review; Knowledge management; Artificial intelligence; MEDLINE; Political science","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.3603913,0.001802459,0.002292903,0.02756288,0.003428323,0.01168751,0.004824677,0.002090786,0.007148868],"category_scores_gemma":[0.7355446,0.001542361,0.004222585,0.01887821,0.003920709,0.008532258,0.0109641,0.002327324,0.001322296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007691759,"about_ca_system_score_gemma":0.01823928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003914285,"about_ca_topic_score_gemma":0.01082193,"domain_scores_codex":[0.6360156,0.297218,0.03568948,0.00904758,0.02067838,0.001350845],"domain_scores_gemma":[0.06586908,0.8702329,0.01898377,0.01950207,0.02460464,0.0008076228],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002695132,0.0007408539,0.03489874,0.1289564,0.004062188,0.00131226,0.1408965,0.01493347,0.006390795,0.03815147,0.01139778,0.6155644],"study_design_scores_gemma":[0.005527645,0.004115047,0.05808738,0.1122167,0.0102264,0.002344689,0.1148216,0.1565818,0.0346836,0.2774495,0.2221199,0.001825784],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2111613,0.02145309,0.6554652,0.01026067,0.0008198234,0.05954476,0.01121534,0.003479436,0.02660048],"genre_scores_gemma":[0.2420966,0.002169832,0.7121408,0.001381803,0.0001284999,0.03802432,0.003020612,0.0002204877,0.0008169885],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6396087,"threshold_uncertainty_score":0.7887514,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2878373345554333,"score_gpt":0.5558607988604336,"score_spread":0.2680234643050003,"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."}}