{"id":"W3217443545","doi":"10.1109/bigdata52589.2021.9672045","title":"MURAL: An Unsupervised Random Forest-Based Embedding for Electronic Health Record Data","year":2021,"lang":"en","type":"preprint","venue":"2021 IEEE International Conference on Big Data (Big Data)","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"National Institutes of Health; Institut de Valorisation des Données; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Categorical variable; Random forest; Embedding; Computer science; Artificial intelligence; Random variable; Dimensionality reduction; Decision tree; Missing data; Data mining; Mathematics; Machine learning; Statistics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science","research_integrity"],"consensus_categories":["open_science"],"category_scores_codex":[0.004524304,0.0008974218,0.001183131,0.0005738587,0.0004251631,0.002699086,0.03966799,0.0004455388,0.0002151939],"category_scores_gemma":[0.00230621,0.000952229,0.0001397458,0.000420675,0.0001113004,0.002353219,0.01801616,0.00258291,0.00004898585],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005728455,"about_ca_system_score_gemma":0.01071417,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005716693,"about_ca_topic_score_gemma":0.02074478,"domain_scores_codex":[0.9884229,0.001237212,0.00155015,0.005705907,0.001808529,0.001275342],"domain_scores_gemma":[0.9745538,0.00104762,0.001251294,0.02158526,0.001038243,0.0005237547],"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":[0.0005233011,0.0004985784,0.0008210879,0.0007946749,0.0004366999,0.00007816117,0.0002225961,0.004848148,0.00006935356,0.0085233,0.03050376,0.9526803],"study_design_scores_gemma":[0.002095494,0.0002712403,0.0002190405,0.001123959,0.00004049721,0.00001735096,0.00007935418,0.9312132,0.00002045107,0.001549031,0.06255797,0.0008124137],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"dataset","genre_scores_codex":[0.001918768,0.0004238914,0.9120944,0.02246203,0.01617965,0.001710505,0.04470114,0.0003055131,0.0002040833],"genre_scores_gemma":[0.3256407,0.001643734,0.09268162,0.003625125,0.004462732,0.0002589476,0.5713372,0.0001392477,0.0002107419],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9518679,"threshold_uncertainty_score":0.9997182,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4029666149370337,"score_gpt":0.4393100574242843,"score_spread":0.03634344248725058,"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."}}