{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005062745,0.00007812545,0.00009739995,0.00003350448,0.0001604054,0.0001281237,0.0004065248,0.0000638849,0.000002694765],"category_scores_gemma":[0.0003556191,0.00007337317,0.00001370719,0.0001819368,0.000005905548,0.0006775464,0.0002815219,0.0001897573,6.335457e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001502868,"about_ca_system_score_gemma":0.00004716348,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001612511,"about_ca_topic_score_gemma":0.000006960784,"domain_scores_codex":[0.9991162,0.00009517998,0.0001040659,0.0004639311,0.00007349896,0.0001471142],"domain_scores_gemma":[0.9992819,0.0002433214,0.00005979762,0.0002696925,0.00005197826,0.00009333796],"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":[0.00002275506,0.000009475135,0.0004925607,0.0001707264,0.0000153455,6.76697e-7,0.001884861,0.06245285,0.004896324,0.008523175,0.008927785,0.9126035],"study_design_scores_gemma":[0.0001026295,0.00005233506,0.0000715306,0.00000584006,0.000004589434,0.00001230939,0.00003869504,0.9934205,0.0003024794,0.0002202369,0.005686914,0.00008189597],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0002472184,0.0001234445,0.9799228,0.01884582,0.0000905212,0.0001613136,0.000002549561,0.000249569,0.0003567328],"genre_scores_gemma":[0.01017084,0.0000094631,0.9882103,0.001304632,0.00009594856,0.00000906787,0.00001203606,0.000008734731,0.0001790432],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9309677,"threshold_uncertainty_score":0.2992071,"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."}}