{"id":"W4411174920","doi":"10.1007/s11192-025-05339-6","title":"Mapping the unseen in practice: comparing latent Dirichlet allocation and BERTopic for navigating topic spaces","year":2025,"lang":"en","type":"article","venue":"Scientometrics","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Montréal; Université de Montréal","funders":"Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung","keywords":"Latent Dirichlet allocation; Topic model; Computer science; Data science; Information retrieval","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["bibliometrics"],"domain":null,"study_design":"simulation_or_modeling","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"simulation_or_modeling","genre":"methods","about_ca_system":false,"about_ca_topic":false,"confidence":"medium","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003968888,0.00004864371,0.000099797,0.0005892284,0.0005238798,0.0002593134,0.0001703364,0.0000303105,0.00000245435],"category_scores_gemma":[0.00410316,0.00003823644,0.00002957564,0.008232093,0.00009853038,0.0001724799,0.0000670994,0.00009618097,8.587571e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001012922,"about_ca_system_score_gemma":0.00008501285,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001098226,"about_ca_topic_score_gemma":0.0006284598,"domain_scores_codex":[0.998965,0.0001820302,0.0001693411,0.0001730572,0.0003374914,0.0001730605],"domain_scores_gemma":[0.9974124,0.002194686,0.00008516003,0.00007353548,0.0002006815,0.00003346661],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00000460698,0.00005373805,0.5176436,0.00003339139,0.00003704091,6.158145e-7,0.0200859,0.001273896,0.0000446103,0.1296709,0.0001469258,0.3310047],"study_design_scores_gemma":[0.0005763288,0.00002073397,0.7455396,0.0001244133,0.00006231874,4.968405e-7,0.02028156,0.06776394,0.00004457236,0.02342787,0.141961,0.0001971257],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.901986,0.001724199,0.05515245,0.02915004,0.0005224028,0.0004612325,8.239328e-7,0.00003080463,0.01097206],"genre_scores_gemma":[0.9881786,0.00008145688,0.0104424,0.0002900588,0.00005612604,0.00001561587,0.000002523215,0.000001646344,0.0009315995],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3308076,"threshold_uncertainty_score":0.4912162,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1049383044834109,"score_gpt":0.4485186821781035,"score_spread":0.3435803776946926,"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."}}