{"id":"W2990325812","doi":"10.23889/ijpds.v4i3.1209","title":"Linked Administrative Data at Statistics Canada – new data resources for horizontal research","year":2019,"lang":"en","type":"article","venue":"International Journal for Population Data Science","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Statistics Canada","funders":"","keywords":"Record linkage; Linkage (software); Immigration; Analytics; Data science; Business; Public relations; Geography; Database; Political science; Sociology; Computer science; Law","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":[0.03257833,0.001593811,0.002349266,0.02741367,0.007304067,0.01525993,0.006810112,0.002869896,0.06013358],"category_scores_gemma":[0.1563664,0.002327951,0.002571478,0.06881136,0.002210141,0.007013549,0.01104749,0.006276311,0.02878895],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.09260286,"about_ca_system_score_gemma":0.302613,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.969501,"about_ca_topic_score_gemma":0.9565055,"domain_scores_codex":[0.9497553,0.007498772,0.004141519,0.003765721,0.03098601,0.003852685],"domain_scores_gemma":[0.6964021,0.03446778,0.009220568,0.04294838,0.1960175,0.0209437],"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":[0.00008294187,0.00003108248,0.004045168,0.0003844073,0.00008894317,0.00007062909,0.0003948926,0.0007659265,0.0001805642,0.0232411,0.9195201,0.05119426],"study_design_scores_gemma":[0.00004506079,0.00000634965,0.00575232,0.0006277739,0.00002833483,0.00002367685,0.0002879066,0.0007732724,0.0002492822,0.0050547,0.9870589,0.00009238276],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.001275568,0.002048709,0.02269964,0.01589875,0.001201191,0.001124249,0.8830137,0.01058629,0.06215192],"genre_scores_gemma":[0.01370508,0.005679494,0.1142741,0.006996979,0.0006441583,0.003005462,0.820311,0.006696693,0.02868696],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9073972,"threshold_uncertainty_score":0.6718839,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6937162468840304,"score_gpt":0.6065499564550645,"score_spread":0.08716629042896595,"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."}}