{"id":"W3113091126","doi":"10.1016/j.bdr.2020.100173","title":"Agreeing to Disagree: Choosing Among Eight Topic-Modeling Methods","year":2020,"lang":"en","type":"article","venue":"Big Data Research","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":31,"is_retracted":false,"has_abstract":false,"ca_institutions":"Simon Fraser University; University of British Columbia","funders":"Hong Kong Polytechnic University","keywords":"Latent Dirichlet allocation; Topic model; Computer science; Data science; Artificial intelligence; Coherence (philosophical gambling strategy); Field (mathematics); Context (archaeology); Machine learning; Management science; Natural language processing","routes":{"ca_aff":true,"ca_fund":false,"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":[0.009259239,0.00009170833,0.0001949661,0.0001888215,0.001004764,0.0003741746,0.001508523,0.00006751506,0.0002103635],"category_scores_gemma":[0.006129509,0.00008455684,0.00005553034,0.002016731,0.0001575994,0.0003724009,0.001405045,0.0003608269,0.0001145059],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000080818,"about_ca_system_score_gemma":0.0002620197,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006251721,"about_ca_topic_score_gemma":0.002660237,"domain_scores_codex":[0.9955417,0.001895229,0.0002518596,0.0005997883,0.001182187,0.0005292619],"domain_scores_gemma":[0.9975135,0.001157119,0.00003124427,0.0005158762,0.0002933542,0.0004888658],"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":[0.00001801186,0.00002496227,0.003656276,0.00001653674,0.00003847262,0.000013008,0.006914508,0.000562519,0.0005418501,0.003426272,0.002598736,0.9821888],"study_design_scores_gemma":[0.0002749028,0.00005755342,0.004460173,0.00007652569,0.00003784292,4.969021e-7,0.006673796,0.5352027,0.0002549113,0.007789413,0.4447511,0.0004205507],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0312003,0.0005318184,0.9240699,0.02678004,0.0002935114,0.0003721925,0.00003846282,0.00007921696,0.01663457],"genre_scores_gemma":[0.8032299,0.0001023059,0.1918783,0.0005724387,0.002881619,0.00001486103,0.0001326033,0.00002024315,0.0011677],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9817683,"threshold_uncertainty_score":0.9450773,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8645336607084751,"score_gpt":0.6295130273678069,"score_spread":0.2350206333406683,"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."}}