{"id":"W4389641434","doi":"10.1017/eds.2023.40","title":"A novel heuristic method for detecting overfit in unsupervised classification of climate model data","year":2023,"lang":"en","type":"article","venue":"Environmental Data Science","topic":"Climate variability and models","field":"Environmental Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Environment Research Council; Sight Research UK; UK Research and Innovation; National Science Foundation","keywords":"Overfitting; Computer science; Cluster analysis; Robustness (evolution); Heuristic; Machine learning; Artificial intelligence; Mixture model; Data mining; Class (philosophy); Gaussian; Artificial neural network","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005006385,0.000549113,0.0009778091,0.004621147,0.0009939602,0.001818968,0.001762215,0.00137181,0.0008813337],"category_scores_gemma":[0.0182286,0.0003420274,0.0006551407,0.00203076,0.001276323,0.0008651756,0.001442997,0.001115391,0.0002423211],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009366669,"about_ca_system_score_gemma":0.001597182,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002848934,"about_ca_topic_score_gemma":0.005128697,"domain_scores_codex":[0.9968897,0.0012186,0.0002278571,0.00051894,0.0009219346,0.0002230289],"domain_scores_gemma":[0.9869568,0.008013341,0.001349895,0.001474148,0.001888617,0.0003172334],"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":[0.0008378582,0.0007471656,0.09683299,0.0003182477,0.0006511695,0.000552812,0.001031399,0.2368203,0.02918435,0.03726481,0.006110413,0.5896484],"study_design_scores_gemma":[0.00004396924,0.00008177595,0.00940733,0.00002461178,0.0000280547,0.0001935134,0.0001243343,0.9751823,0.005244773,0.008448679,0.001178049,0.00004254214],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1219828,0.0001259653,0.8754092,0.0001479981,0.0000326434,0.0001218523,0.0001798671,0.0007954481,0.001204176],"genre_scores_gemma":[0.5062557,0.00003970979,0.4922014,0.0001232159,0.00004336764,0.0001826453,0.0005688509,0.0001212661,0.0004638351],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005006385,"threshold_uncertainty_score":0.02647662,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1606655707973861,"score_gpt":0.3548466895345715,"score_spread":0.1941811187371854,"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."}}