{"id":"W4283213951","doi":"10.1109/iemtronics55184.2022.9795745","title":"Intelligent Feature Selection on Multivariate Dataset using Advanced Data Profiling","year":2022,"lang":"en","type":"article","venue":"2022 IEEE International IOT, Electronics and Mechatronics Conference (IEMTRONICS)","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Feature selection; Profiling (computer programming); Computer science; Data mining; Machine learning; Multivariate statistics; Artificial intelligence","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":[0.001713757,0.0008207612,0.001068614,0.003543458,0.0004325707,0.001151218,0.0005714319,0.0005175006,0.0006557635],"category_scores_gemma":[0.005444457,0.0001613492,0.001038135,0.002734379,0.0002289786,0.0008234677,0.0007678106,0.000689309,0.0004193905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004081464,"about_ca_system_score_gemma":0.0005449952,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001720334,"about_ca_topic_score_gemma":0.001567581,"domain_scores_codex":[0.9988814,0.0003227504,0.000141711,0.0002716834,0.0002730611,0.0001093918],"domain_scores_gemma":[0.997649,0.001341972,0.0002498337,0.0003087371,0.0003746366,0.00007572836],"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":[0.000895447,0.0005338565,0.04441261,0.0003055854,0.0003263172,0.000743539,0.000250382,0.1069177,0.03748658,0.001923923,0.008749937,0.7974541],"study_design_scores_gemma":[0.00005083231,0.0002704055,0.02028833,0.00004070694,0.00009937394,0.0003668929,0.0001774955,0.9479436,0.02059541,0.00540742,0.004710342,0.00004920812],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2746464,0.0006871417,0.7121168,0.0006548328,0.0001149447,0.0003563634,0.004838006,0.005668995,0.0009164725],"genre_scores_gemma":[0.6542778,0.000274223,0.3359085,0.00009585732,0.00007740327,0.0003784771,0.008264754,0.00008987888,0.0006331204],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003543458,"threshold_uncertainty_score":0.009063303,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06475998024089975,"score_gpt":0.3384563381554218,"score_spread":0.2736963579145221,"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."}}