{"id":"W4220828579","doi":"10.1038/s41597-022-01197-6","title":"An expert-curated global database of online newspaper articles on spiders and spider bites","year":2022,"lang":"en","type":"article","venue":"Scientific Data","topic":"Animal and Plant Science Education","field":"Psychology","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"The Scarborough Hospital; McGill University; University of Toronto; Espace pour la vie","funders":"Science and Engineering Research Board; Helsingin Yliopisto","keywords":"Spider; Newspaper; Metadata; Social media; Data science; Database; World Wide Web; Computer science; Ecology; Biology; Sociology; Media studies","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003214733,0.001289708,0.001438609,0.0447074,0.0009773616,0.002840748,0.001108551,0.001194396,0.02273214],"category_scores_gemma":[0.02108625,0.0005590562,0.0007101588,0.03347217,0.0007383259,0.002110491,0.002363978,0.00102645,0.0174713],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001000063,"about_ca_system_score_gemma":0.003973141,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005931899,"about_ca_topic_score_gemma":0.01343148,"domain_scores_codex":[0.9959235,0.0005798987,0.00129489,0.0008436507,0.001087274,0.0002707097],"domain_scores_gemma":[0.960609,0.01953926,0.006712587,0.002633601,0.008838284,0.001667363],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001100896,0.0005773981,0.04779238,0.05039156,0.0005010625,0.003093603,0.0065605,0.0009493406,0.01306866,0.003210847,0.6480381,0.2247156],"study_design_scores_gemma":[0.0001325446,0.0001409382,0.1788633,0.00406148,0.0004335124,0.00119348,0.002512556,0.0007512484,0.003294767,0.0008156357,0.8076679,0.0001326283],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.01687385,0.001878178,0.001416133,0.0002178532,0.00009870007,0.0006402535,0.9720659,0.0006824504,0.006126569],"genre_scores_gemma":[0.01899655,0.001830331,0.01079441,0.0002008538,0.0001668971,0.001933394,0.9628733,0.0002859831,0.002918199],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.0447074,"threshold_uncertainty_score":0.07604659,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1299222449670548,"score_gpt":0.3950524839355657,"score_spread":0.265130238968511,"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."}}