{"id":"W4235282642","doi":"10.21203/rs.3.rs-40780/v2","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":"","keywords":"Relevance (law); Decoding methods; Information retrieval; Computer science; Data science; Algorithm; Political science","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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2326506,0.001524518,0.002448966,0.006215493,0.0006355573,0.003254774,0.001788639,0.001167142,0.002559617],"category_scores_gemma":[0.639156,0.001424259,0.006765517,0.00729859,0.001081927,0.00348898,0.002365329,0.001493996,0.001187167],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001778179,"about_ca_system_score_gemma":0.005148071,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002113311,"about_ca_topic_score_gemma":0.003488216,"domain_scores_codex":[0.7338392,0.1860015,0.04984994,0.01238151,0.01669098,0.001236885],"domain_scores_gemma":[0.1539866,0.7154519,0.08192882,0.03017897,0.01764275,0.0008108116],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.009649062,0.0002475765,0.5755705,0.03581409,0.01528169,0.00134781,0.002737292,0.02290622,0.004313636,0.001916495,0.01665531,0.3135603],"study_design_scores_gemma":[0.005240642,0.007547266,0.5412112,0.02561759,0.06440071,0.01091846,0.001913532,0.2291922,0.025456,0.02346116,0.06375556,0.001285761],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6317617,0.05205441,0.2523855,0.005601855,0.0006803422,0.006449043,0.0442127,0.003227203,0.003627315],"genre_scores_gemma":[0.8945676,0.003115374,0.08756138,0.001032999,0.0001833783,0.003659631,0.009143826,0.0003796717,0.0003562243],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7673494,"threshold_uncertainty_score":0.9462785,"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."}}