{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2672302,0.001512229,0.00287775,0.008844007,0.0007294773,0.003983405,0.001953626,0.001321071,0.002998283],"category_scores_gemma":[0.6949728,0.001483674,0.006523164,0.00873395,0.001283972,0.004114172,0.00266619,0.001485297,0.001273293],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00169938,"about_ca_system_score_gemma":0.005196059,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001824192,"about_ca_topic_score_gemma":0.00326766,"domain_scores_codex":[0.6496381,0.2099635,0.09288155,0.01763498,0.02817329,0.001708754],"domain_scores_gemma":[0.108087,0.7339739,0.1009031,0.03342448,0.0227838,0.0008276891],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.008723616,0.000184353,0.6092747,0.05650199,0.01769847,0.001386685,0.003345305,0.01083323,0.003604951,0.001889366,0.01749425,0.269063],"study_design_scores_gemma":[0.003783457,0.005490696,0.657726,0.04334126,0.06762097,0.01179306,0.001988892,0.09643535,0.02129139,0.01752924,0.07181421,0.00118551],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6126906,0.0926225,0.2094615,0.006998067,0.001024466,0.009567576,0.05898556,0.003147761,0.00550191],"genre_scores_gemma":[0.9096271,0.004742776,0.06911783,0.001135699,0.0002983528,0.005000511,0.009157445,0.0004586359,0.000461741],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7327698,"threshold_uncertainty_score":0.9036357,"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."}}