{"id":"W6977706616","doi":"10.7910/dvn/sgw3ps","title":"Supplementary material - A Bayesian latent class analysis approach for assessing the accuracy of natural language processing models classifying topic and sentiment of public health-related social media content","year":2025,"lang":"en","type":"dataset","venue":"Harvard Dataverse","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Latent class model; Bayesian probability; Topic model; Class (philosophy); Public health; Naive Bayes classifier; Bayesian inference","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003636004,0.001461923,0.0005816244,0.003293044,0.001011486,0.001798461,0.002420999,0.001844294,0.1349914],"category_scores_gemma":[0.02152776,0.0005756404,0.0009654941,0.003672673,0.0005451896,0.0009992617,0.001602746,0.001310541,0.06272568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002948856,"about_ca_system_score_gemma":0.002932237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05360453,"about_ca_topic_score_gemma":0.1336775,"domain_scores_codex":[0.9983884,0.0004757564,0.0001806352,0.0003393764,0.0004532438,0.000162617],"domain_scores_gemma":[0.9849505,0.008415423,0.0008283509,0.002460172,0.002865798,0.0004797164],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001701972,0.0001234923,0.005788171,0.0004945937,0.00006175302,0.00004437489,0.00006300475,0.001798973,0.0002009447,0.00179177,0.9819723,0.00749055],"study_design_scores_gemma":[0.001067822,0.0001621501,0.03006814,0.0006310762,0.0001061583,0.0002434083,0.0003588806,0.01275692,0.002197752,0.01211384,0.9401786,0.0001152141],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0005642496,0.00002716571,0.0004290706,0.00009335209,0.00002335336,0.0000373254,0.9978552,0.0003158789,0.0006543981],"genre_scores_gemma":[0.001882982,0.00001884495,0.001957374,0.00005068295,0.000008843458,0.000239175,0.9949765,0.00006501907,0.000800562],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.1349914,"threshold_uncertainty_score":0.4515914,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07107797139428218,"score_gpt":0.3225322191671376,"score_spread":0.2514542477728554,"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."}}