{"id":"W4405765420","doi":"10.48550/arxiv.2412.16209","title":"Challenges in the calibration of tree-based models for imbalanced classification","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Hyperparameter; Tree (set theory); Machine learning; Artificial intelligence; Statistics; Econometrics; Random forest; Computer science; Mathematics; Psychology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003391302,0.0001806739,0.0003367428,0.0002269483,0.00002245349,0.00001358023,0.0003141646,0.0001899112,0.000006530826],"category_scores_gemma":[0.00005492739,0.0001592468,0.000197877,0.0002619348,0.00007983358,0.00006885581,0.0001205268,0.0002928983,0.000004367054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001345107,"about_ca_system_score_gemma":0.0002952852,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000337643,"about_ca_topic_score_gemma":0.0002273672,"domain_scores_codex":[0.9987965,0.0001243555,0.0002223981,0.0006048167,0.00009612441,0.0001558123],"domain_scores_gemma":[0.9986345,0.0001739545,0.0001764727,0.0008386322,0.0001227304,0.00005376698],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003494406,0.00145657,0.01244487,0.009615009,0.0005123887,0.0002947923,0.001510157,0.5099536,0.001384733,0.4468546,0.002100912,0.01037805],"study_design_scores_gemma":[0.001073473,0.0000792076,0.01239995,0.0004017963,0.0002103077,6.251334e-7,0.0003586691,0.9452165,0.0001189556,0.03982884,0.0001578221,0.0001538333],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8202553,0.002694472,0.1444831,0.005348729,0.0006251074,0.005225268,0.001645131,0.0004191696,0.01930373],"genre_scores_gemma":[0.9983619,0.0004972877,0.0001872649,0.00007078778,0.0000597637,0.00001458481,0.0006505112,0.00002265099,0.0001351942],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.435263,"threshold_uncertainty_score":0.6493894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2207264365983063,"score_gpt":0.2518108476678467,"score_spread":0.03108441106954041,"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."}}