{"id":"W6931251810","doi":"10.5281/zenodo.4000748","title":"Draft template for data extraction","year":2020,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Extraction (chemistry); Data extraction; Pattern recognition (psychology); Hull; Matching (statistics)","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":[0.01014193,0.00189958,0.001812469,0.007796603,0.001363285,0.004655293,0.002932896,0.002272991,0.2367109],"category_scores_gemma":[0.08167681,0.002350903,0.002804476,0.0058978,0.0007639577,0.003382938,0.004287286,0.003005207,0.1984666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001757945,"about_ca_system_score_gemma":0.009322864,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004359382,"about_ca_topic_score_gemma":0.004555763,"domain_scores_codex":[0.9910011,0.002235598,0.003038694,0.0009463563,0.002382975,0.0003953832],"domain_scores_gemma":[0.9133249,0.03533363,0.002526792,0.01632245,0.03108816,0.001404059],"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":[0.0003380951,0.0001083921,0.0008594982,0.003350035,0.00009451547,0.0003961609,0.0005870175,0.0008979226,0.007082272,0.008383255,0.8332201,0.1446826],"study_design_scores_gemma":[0.0001397305,0.00005240812,0.0008611436,0.000726477,0.00006831249,0.0002899394,0.0001921565,0.001268585,0.006272336,0.008362089,0.9817113,0.00005564793],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.001363733,0.0005955798,0.4610647,0.003132651,0.002292588,0.01327617,0.4460695,0.03279671,0.03940845],"genre_scores_gemma":[0.005088894,0.0007543438,0.5452718,0.002070446,0.0004338459,0.01934289,0.3723986,0.01194173,0.04269748],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.2367109,"threshold_uncertainty_score":0.7918768,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1330825545238107,"score_gpt":0.2989105927735164,"score_spread":0.1658280382497057,"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."}}