{"id":"W2768421170","doi":"10.1016/j.ins.2017.11.041","title":"Automatic feature engineering for regression models with machine learning: An evolutionary computation and statistics hybrid","year":2017,"lang":"en","type":"article","venue":"Information Sciences","topic":"Evolutionary Algorithms and Applications","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Conselho Nacional de Desenvolvimento Científico e Tecnológico; Natural Sciences and Engineering Research Council of Canada; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; Universal","keywords":"Genetic programming; Computer science; Feature engineering; Symbolic regression; Interpretability; Machine learning; Artificial intelligence; Randomness; Feature (linguistics); Benchmark (surveying); Evolutionary computation; Regression; Data mining; Statistics; Mathematics; Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00170378,0.0005712904,0.001156368,0.001034314,0.0004456195,0.001076729,0.001107119,0.001175678,0.001666205],"category_scores_gemma":[0.006328037,0.0006051176,0.001133889,0.0009578536,0.0006065212,0.001569369,0.001189595,0.001391278,0.000409467],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006952159,"about_ca_system_score_gemma":0.0009959722,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002194921,"about_ca_topic_score_gemma":0.002869749,"domain_scores_codex":[0.9993023,0.0003056094,0.00004216334,0.0001155122,0.0001917783,0.00004267354],"domain_scores_gemma":[0.9980555,0.001309232,0.0001187938,0.0002327154,0.0002484563,0.00003541483],"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":[0.00008000044,0.000147529,0.001298868,0.0001271522,0.0001165654,0.0001019886,0.0000720668,0.5431708,0.005452842,0.07121982,0.001840879,0.3763714],"study_design_scores_gemma":[0.000003570668,0.00001445579,0.00009230944,0.000004036361,0.000005771458,0.00001299569,0.000002394459,0.9884959,0.0003991601,0.01064322,0.0003223256,0.000003700909],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00465251,0.0001390615,0.994422,0.000112568,0.00001988297,0.00001381407,0.0000123817,0.0001801397,0.0004476964],"genre_scores_gemma":[0.2808248,0.0003123233,0.7152293,0.0001304143,0.0001086034,0.0001557476,0.0001112219,0.0002440443,0.002883514],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002194921,"threshold_uncertainty_score":0.009010553,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0199954237668725,"score_gpt":0.2768263841359542,"score_spread":0.2568309603690817,"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."}}