{"id":"W4309221281","doi":"10.21203/rs.3.rs-2260181/v1","title":"Large Scale Genealogical Information Extraction From Handwritten Quebec Parish Records","year":2022,"lang":"en","type":"preprint","venue":"Research Square","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Chicoutimi","funders":"Association Nationale de la Recherche et de la Technologie","keywords":"Workflow; Computer science; Consistency (knowledge bases); Scale (ratio); Sample (material); Information extraction; Artificial intelligence; Population; Natural language processing; Information retrieval; Database; Data mining; Geography; Cartography; Medicine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.0005308702,0.0008749057,0.0005146759,0.008809743,0.001438227,0.002165502,0.0008486236,0.0005930578,0.007358922],"category_scores_gemma":[0.003309823,0.0002827686,0.0005029967,0.005422539,0.0004769292,0.0005403198,0.0006990293,0.0004096803,0.003816564],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003337497,"about_ca_system_score_gemma":0.006565708,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.6058118,"about_ca_topic_score_gemma":0.6809731,"domain_scores_codex":[0.9992261,0.00007044485,0.0000492994,0.00024196,0.0003048375,0.0001074211],"domain_scores_gemma":[0.9972565,0.0007073225,0.0001960209,0.0003798166,0.001326624,0.0001337919],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003344079,0.0001533686,0.03765109,0.0009795934,0.0001854034,0.002585572,0.002689046,0.01294063,0.04752021,0.003142668,0.09737046,0.7944475],"study_design_scores_gemma":[0.000082189,0.0001055028,0.2897139,0.000557541,0.0001847488,0.001225579,0.004346179,0.2024248,0.08493982,0.004530121,0.4116317,0.0002578573],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.3858991,0.002816879,0.2790208,0.001782406,0.0002572962,0.002219479,0.2473726,0.04789256,0.03273888],"genre_scores_gemma":[0.3601589,0.0009985087,0.3921121,0.0002021707,0.00008881139,0.0007466423,0.2041506,0.001222329,0.04031989],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3941882,"threshold_uncertainty_score":0.7930192,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1050366162942268,"score_gpt":0.4819845324814767,"score_spread":0.3769479161872499,"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."}}