{"id":"W4395112800","doi":"10.1038/s41597-024-03275-3","title":"A new commercial boundary dataset for metropolitan areas in the USA and Canada, built from open data","year":2024,"lang":"en","type":"article","venue":"Scientific Data","topic":"Human Mobility and Location-Based Analysis","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"University of Toronto","keywords":"Metropolitan area; Boundary (topology); Geography; Regional science; Cluster analysis; Data collection; Data science; Segmentation; Economic geography; Cartography; Computer science; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003685466,0.0007961467,0.0007767107,0.006462654,0.002146125,0.001803027,0.001919785,0.0007326524,0.004966723],"category_scores_gemma":[0.003807909,0.0003741099,0.0008394935,0.01561281,0.000610384,0.001066199,0.002253637,0.001066128,0.00332236],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.009385867,"about_ca_system_score_gemma":0.01895326,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.9607536,"about_ca_topic_score_gemma":0.9803773,"domain_scores_codex":[0.9992149,0.0000406635,0.00005936479,0.0001909215,0.0003153202,0.0001788913],"domain_scores_gemma":[0.9974396,0.0001757385,0.000185933,0.0002733546,0.001580857,0.0003444224],"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":[0.0002960811,0.0001743989,0.1099896,0.001048137,0.0002285733,0.0005255889,0.001880207,0.008077192,0.001451046,0.006981876,0.8113893,0.05795796],"study_design_scores_gemma":[0.0001235945,0.00002468018,0.3078751,0.0005017336,0.00009484431,0.000328129,0.004557607,0.0166084,0.00192988,0.003346439,0.6644447,0.0001648181],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0232518,0.0002835864,0.001956497,0.0001547495,0.00002857804,0.000111661,0.9662409,0.001100863,0.006871298],"genre_scores_gemma":[0.02741274,0.0001640809,0.00484369,0.00004491489,0.000006866599,0.0001777425,0.9658884,0.0001403894,0.001321139],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03924638,"threshold_uncertainty_score":0.07895499,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1611647706221825,"score_gpt":0.4182198925424139,"score_spread":0.2570551219202314,"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."}}