{"id":"W4231631488","doi":"10.1002/sam.10048","title":"Application of model selection technique in chemogenomic data analysis","year":2009,"lang":"en","type":"article","venue":"Statistical Analysis and Data Mining The ASA Data Science Journal","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; York University","funders":"National Cancer Institute; Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Data mining; Dependency (UML); Model selection; Bayesian network; Selection (genetic algorithm); Data set; Construct (python library); Set (abstract data type); Bayesian information criterion; Bayesian probability; Artificial intelligence; Machine learning","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01669154,0.001550059,0.002084939,0.003906617,0.001227264,0.00142708,0.001483359,0.00102013,0.001331148],"category_scores_gemma":[0.03416061,0.0006115912,0.002560943,0.002965866,0.0009698569,0.0008478985,0.00123757,0.002442319,0.0003441407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001052192,"about_ca_system_score_gemma":0.002605186,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004244085,"about_ca_topic_score_gemma":0.00397052,"domain_scores_codex":[0.9877139,0.01009053,0.0003820655,0.0007604462,0.0008701402,0.0001828517],"domain_scores_gemma":[0.9651266,0.03144674,0.0009677343,0.0009677674,0.001223138,0.0002680766],"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":[0.0005951691,0.0003544022,0.02180018,0.000391453,0.002774166,0.0009023994,0.0003238411,0.7293355,0.005443244,0.02333119,0.004722021,0.2100265],"study_design_scores_gemma":[0.0000322515,0.00006228442,0.0007565831,0.00001134962,0.00006550119,0.0000570397,0.00001666766,0.9870194,0.0007168549,0.01082207,0.0004250959,0.00001488815],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0132766,0.0001909797,0.9851595,0.0003052313,0.00002983471,0.0001370176,0.0001846344,0.0005529424,0.0001633622],"genre_scores_gemma":[0.4114432,0.0002831145,0.5850589,0.0002276042,0.0001053868,0.0009694022,0.001222487,0.0001688936,0.0005210101],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01669154,"threshold_uncertainty_score":0.08827442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04998375380142114,"score_gpt":0.3711030331056732,"score_spread":0.321119279304252,"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."}}