{"id":"W4378232119","doi":"10.1093/jssam/smad015","title":"Automated Classification for Open-Ended Questions with BERT","year":2023,"lang":"en","type":"article","venue":"Journal of Survey Statistics and Methodology","topic":"Topic Modeling","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Western University","funders":"Social Sciences and Humanities Research Council; Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Coding (social sciences); Artificial intelligence; Boosting (machine learning); Natural language processing; Machine learning; Language model; Training set; Statistics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005758173,0.0000604877,0.0002199179,0.0001277735,0.00008284271,0.000086765,0.0003399874,0.00004289773,8.999509e-7],"category_scores_gemma":[0.001906952,0.00004608402,0.00001200248,0.0002084297,0.00003401298,0.0001483375,0.00008416952,0.0000900076,7.339621e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001511731,"about_ca_system_score_gemma":0.0001271321,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009544994,"about_ca_topic_score_gemma":0.0001424311,"domain_scores_codex":[0.9985463,0.000786465,0.0003043692,0.0001352718,0.00009661967,0.0001310008],"domain_scores_gemma":[0.9952788,0.003793358,0.0002789434,0.0001402677,0.0004403151,0.00006827304],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003361366,0.0001042936,0.01437354,0.00008335959,0.000237161,0.00007285874,0.001278574,0.002523723,0.003472308,0.7300373,0.03569883,0.2117819],"study_design_scores_gemma":[0.0007217293,0.0003719457,0.3934684,0.00001777633,0.00001978638,0.0001147884,0.00004077257,0.5618527,0.00004669099,0.04184914,0.001403324,0.00009299866],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004412487,0.00003078412,0.9939578,0.001046335,0.0003372378,0.0001069125,0.00005317494,0.00003606223,0.00001922265],"genre_scores_gemma":[0.02782387,0.00006616203,0.9718629,0.0001074984,0.00002779667,0.000005253576,0.00001215475,0.000005473625,0.00008894091],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6881881,"threshold_uncertainty_score":0.2282937,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4858752837440197,"score_gpt":0.4643715025409455,"score_spread":0.02150378120307417,"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."}}