{"id":"W3162439607","doi":"10.31234/osf.io/9a52q","title":"Content matters: Measures of contextual diversity must consider semantic content","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Optimal distinctiveness theory; Diversity (politics); Word (group theory); Computer science; Natural language processing; Word lists by frequency; Context (archaeology); Artificial intelligence; Variance (accounting); Linguistics; Content (measure theory); Psychology; Mathematics; Social psychology; Sociology; History","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":["metaepi_narrow","open_science"],"consensus_categories":[],"category_scores_codex":[0.0005201449,0.0003639947,0.0008303503,0.000143128,0.0001295426,0.0002408013,0.001618836,0.0002523072,0.0001341304],"category_scores_gemma":[0.0001351996,0.000335801,0.0003672772,0.00009132911,0.0001514682,0.0002408939,0.01016802,0.00050768,0.00001792862],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001187106,"about_ca_system_score_gemma":0.0002702367,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005550976,"about_ca_topic_score_gemma":0.0005367441,"domain_scores_codex":[0.9969217,0.0002082278,0.0007035552,0.0009802846,0.0008123721,0.0003738978],"domain_scores_gemma":[0.9969248,0.0001855774,0.0003782529,0.001544628,0.0008060037,0.0001607482],"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":[0.0003343268,0.004488159,0.2165139,0.00874665,0.01201892,0.004355304,0.1194039,0.02548334,0.0540166,0.3613201,0.06312929,0.1301894],"study_design_scores_gemma":[0.01886566,0.0008272658,0.09905808,0.01100443,0.001884374,0.0008096294,0.06070828,0.6573301,0.1161171,0.01828384,0.003352911,0.01175838],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1644694,0.0006885427,0.8273053,0.004342117,0.001652148,0.0004718598,0.00001273234,0.0001760127,0.0008818854],"genre_scores_gemma":[0.9724558,0.00004912774,0.02376238,0.003045984,0.00005439845,0.00001175058,0.000008291086,0.00001256276,0.0005997107],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8079864,"threshold_uncertainty_score":0.9999094,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2554150969675804,"score_gpt":0.2713124592966478,"score_spread":0.01589736232906747,"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."}}