{"id":"W3139141769","doi":"10.1109/imcom51814.2021.9377412","title":"Natural Sciences Meet Social Sciences: Census Data Analytics for Detecting Home Language Shifts","year":2021,"lang":"en","type":"article","venue":"","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Manitoba","funders":"Natural Sciences and Engineering Research Council of Canada; University of Manitoba","keywords":"Census; Microdata (statistics); Analytics; Computer science; Geography; Data science; Population; Sociology; Demography","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":[0.003788307,0.0009952598,0.0008701982,0.007912579,0.0006346246,0.002073918,0.00104446,0.0008519272,0.008138681],"category_scores_gemma":[0.03530846,0.0005436479,0.0007497258,0.01312074,0.0002517003,0.002206696,0.001914176,0.001419865,0.006370811],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001133713,"about_ca_system_score_gemma":0.003262169,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02804086,"about_ca_topic_score_gemma":0.0290892,"domain_scores_codex":[0.99699,0.0005973507,0.0004890835,0.0004902565,0.001292115,0.0001411627],"domain_scores_gemma":[0.9846404,0.004300504,0.002718892,0.003029737,0.004148019,0.001162455],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.0004314434,0.0004271661,0.174054,0.001769357,0.0006529295,0.0003975206,0.0008485956,0.00920427,0.002104521,0.0148721,0.5068915,0.2883466],"study_design_scores_gemma":[0.0002593833,0.0002026296,0.2689023,0.0006865167,0.0001816689,0.0004362505,0.002331233,0.2009981,0.006709357,0.05582445,0.4632635,0.0002046836],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.04074141,0.001679324,0.07172612,0.004899181,0.0003663535,0.001700419,0.840695,0.020542,0.01765036],"genre_scores_gemma":[0.2171478,0.002134278,0.2045352,0.001023092,0.0003715416,0.003263603,0.5655099,0.001262933,0.00475174],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02804086,"threshold_uncertainty_score":0.05575526,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1141168266916986,"score_gpt":0.3677779799790345,"score_spread":0.2536611532873359,"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."}}