{"id":"W3108384885","doi":"10.21203/rs.3.rs-40780/v1","title":"Decoding semi-automated title-abstract screening: a retrospective exploration of the review, study, and publication characteristics associated with accurate relevance predictions","year":2020,"lang":"en","type":"preprint","venue":"Research Square","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Government of Canada; Agency for Healthcare Research and Quality; U.S. Department of Health and Human Services","keywords":"Relevance (law); Decoding methods; Computer science; Information retrieval; Data science; Data mining; Algorithm; Political science","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0005860442,0.0000989684,0.0001749003,0.0000394395,0.0001044841,0.00003888629,0.0001877582,0.0001704161,0.00001195221],"category_scores_gemma":[0.004672136,0.00006900662,0.00003616978,0.0002469744,0.0001483841,0.000005747756,0.0003389106,0.0004605144,0.000002327347],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000330618,"about_ca_system_score_gemma":0.0001540833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001088769,"about_ca_topic_score_gemma":0.00001669188,"domain_scores_codex":[0.998818,0.0001984249,0.0002027666,0.0003021891,0.0003444833,0.000134181],"domain_scores_gemma":[0.998787,0.00006175748,0.0002059207,0.0003084376,0.0005880275,0.00004880095],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.001018167,0.002637183,0.2958436,0.01391089,0.003895407,0.00005790592,0.004332189,0.0001528547,0.0437145,0.0003883553,0.5205821,0.1134669],"study_design_scores_gemma":[0.0006515341,0.001117711,0.9689682,0.007003869,0.0001250402,0.000003967467,0.0007346782,0.006196687,0.001535755,0.0001804562,0.0131045,0.0003776347],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9313177,0.01668629,0.01098459,0.02058203,0.0007116767,0.009119485,0.002337222,0.0007896203,0.007471364],"genre_scores_gemma":[0.9963923,0.002732961,0.0001099464,0.00002996102,0.00007026159,0.00009968509,0.0003949727,0.0000140545,0.0001559219],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6731246,"threshold_uncertainty_score":0.5593321,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.107370566133873,"score_gpt":0.3986195865911361,"score_spread":0.2912490204572631,"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."}}