{"id":"W4389387146","doi":"10.3233/sji-230063","title":"Classifying respondent comments from the 2021 Canadian Census of Population using machine learning methods1","year":2023,"lang":"en","type":"article","venue":"Statistical Journal of the IAOS","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Respondent; Census; Computer science; Categorization; Encoder; Population; Artificial intelligence; Transformer; Machine learning; Natural language processing; Statistics; Econometrics; Geography; Demography; Mathematics; Sociology; Engineering; Political science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001246813,0.0000733804,0.0001611687,0.0000846298,0.0002295622,0.00006910823,0.0006786428,0.0000334731,0.00002396127],"category_scores_gemma":[0.00103758,0.00004392289,0.00006082859,0.000290274,0.00003044587,0.00009738321,0.0001916253,0.0003849749,0.000002263004],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001484138,"about_ca_system_score_gemma":0.0001433295,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03676289,"about_ca_topic_score_gemma":0.004233847,"domain_scores_codex":[0.9981605,0.0006892446,0.0004085539,0.0001067271,0.0004372633,0.0001977482],"domain_scores_gemma":[0.9981663,0.001102804,0.000286199,0.0002467664,0.00008989507,0.000108047],"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":[0.0001792626,0.00008219406,0.2188438,0.00005840221,0.0005037489,0.0005205275,0.006266409,0.2490497,0.01397026,0.1420509,0.002807107,0.3656677],"study_design_scores_gemma":[0.0002316894,0.00002359555,0.06015402,0.0001038054,0.00003216359,0.00002439055,0.0001148714,0.9164231,0.000145118,0.02189548,0.0007847651,0.00006701644],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1387494,0.00009387655,0.8548819,0.005221307,0.0008694027,0.00007897679,0.00006102384,0.000007326094,0.00003685502],"genre_scores_gemma":[0.8136553,0.000007467381,0.1860349,0.0001648345,0.0001014967,2.715009e-7,0.000003840591,0.000006660959,0.00002528502],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6749059,"threshold_uncertainty_score":0.9696514,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09543310328496614,"score_gpt":0.3544916148562567,"score_spread":0.2590585115712906,"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."}}