{"id":"W3040021588","doi":"10.1145/3406601.3406602","title":"Data Mining Methods for Optimizing Feature Extraction and Model Selection","year":2020,"lang":"en","type":"article","venue":"","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Microsoft (Canada)","funders":"","keywords":"Data mining; Computer science; Context (archaeology); Feature selection; Feature extraction; Data extraction; Feature (linguistics); Data modeling; On the fly; Machine learning; Artificial intelligence; Database","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":[0.005759107,0.002732912,0.002062604,0.002545236,0.0006350039,0.001489008,0.001851295,0.001217333,0.002678056],"category_scores_gemma":[0.02070211,0.0009011375,0.001617286,0.002720902,0.0005445653,0.002045273,0.001209914,0.002155823,0.001416707],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008939327,"about_ca_system_score_gemma":0.001908746,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003631405,"about_ca_topic_score_gemma":0.004496367,"domain_scores_codex":[0.9978114,0.001059436,0.0002276173,0.0003919601,0.0004112989,0.00009820689],"domain_scores_gemma":[0.991079,0.006898385,0.0004986859,0.0006384619,0.0008154364,0.0000700274],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002473748,0.0003849277,0.005025454,0.0004569482,0.0005154944,0.0001838989,0.0001116749,0.4022036,0.005174668,0.01005175,0.009798489,0.5658457],"study_design_scores_gemma":[0.0000677588,0.00006316747,0.0006792899,0.0000369906,0.00006538573,0.00005814385,0.0000323394,0.9791684,0.001628335,0.01673632,0.001446765,0.00001700829],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007884209,0.000652445,0.9886304,0.000560457,0.00004311811,0.0001645673,0.0003277478,0.001232972,0.0005040117],"genre_scores_gemma":[0.1424932,0.0008003969,0.8524778,0.0004370653,0.0001396381,0.0009073344,0.001525157,0.0002985187,0.0009207716],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005759107,"threshold_uncertainty_score":0.03045744,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1621687686099375,"score_gpt":0.4577575793401237,"score_spread":0.2955888107301862,"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."}}