{"id":"W3209399758","doi":"10.1136/oem-2021-epi.442","title":"S-464 Automated Occupational Encoding to the Canadian National Occupation Classification using an Ensemble Classifier from TF-IDF and Doc2Vec Embeddings","year":2021,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick","funders":"","keywords":"Computer science; Coding (social sciences); Classifier (UML); Support vector machine; Data mining; Artificial intelligence; Machine learning","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":[0.0004234056,0.001103465,0.0004507801,0.001822643,0.0009602344,0.001160344,0.0008099315,0.0006356806,0.01348145],"category_scores_gemma":[0.002145075,0.0001901961,0.0008081868,0.001807447,0.0002763349,0.0006406991,0.0006961507,0.0008626703,0.008157155],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002748579,"about_ca_system_score_gemma":0.00607711,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.4249875,"about_ca_topic_score_gemma":0.5199836,"domain_scores_codex":[0.9994066,0.00004228259,0.00002561599,0.0001397169,0.0002374361,0.000148389],"domain_scores_gemma":[0.999069,0.00009621686,0.00002306977,0.00009352431,0.0006538418,0.00006429064],"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":[0.0006775954,0.0002392196,0.01909692,0.0002355956,0.00009952692,0.0004327943,0.0003389706,0.01665059,0.02756133,0.002362942,0.1410978,0.7912068],"study_design_scores_gemma":[0.00006817006,0.0002603355,0.04695931,0.0001308645,0.0001148189,0.0003236857,0.001432065,0.7984725,0.05986404,0.002913314,0.08933354,0.0001273414],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.526192,0.001589415,0.2821368,0.002211132,0.002826489,0.001138946,0.0887647,0.04825897,0.04688149],"genre_scores_gemma":[0.5469954,0.0006316298,0.2346918,0.0004331807,0.0002373489,0.0005666446,0.1481746,0.001054449,0.06721509],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5750124,"threshold_uncertainty_score":0.8450276,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3473811814956081,"score_gpt":0.5158195507423838,"score_spread":0.1684383692467757,"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."}}