{"id":"W4405488355","doi":"10.36227/techrxiv.173143474.47669751/v2","title":"Exploring the Preprocessing of a Time-Series Healthcare Administrative Dataset on Deep Learning to Improve Prediction","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Impact of AI and Big Data on Business and Society","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hotchkiss Brain Institute; University of Calgary","funders":"University of Calgary","keywords":"Preprocessor; Computer science; Time series; Series (stratigraphy); Health care; Artificial intelligence; Deep learning; Data pre-processing; Machine learning; Data mining; Data science; Political science","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":[0.003543948,0.001172372,0.0004803425,0.001187361,0.0003662502,0.001374068,0.001070531,0.0007274023,0.001906178],"category_scores_gemma":[0.01433979,0.0002803291,0.0008928602,0.001521548,0.0003687823,0.001803272,0.0008126182,0.002470153,0.000946786],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009118049,"about_ca_system_score_gemma":0.001694822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009809363,"about_ca_topic_score_gemma":0.01306141,"domain_scores_codex":[0.9989237,0.0004627465,0.0001036705,0.0002013214,0.0002068138,0.000101662],"domain_scores_gemma":[0.9946755,0.003269984,0.0003215092,0.0008013624,0.000824237,0.0001075162],"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":[0.0005224131,0.001005605,0.03612558,0.0003480468,0.0003275671,0.0003093698,0.0002340576,0.4229374,0.01274434,0.008512912,0.01994427,0.4969883],"study_design_scores_gemma":[0.0000196785,0.0001193753,0.005247276,0.00003504281,0.00001997452,0.00003014116,0.00006736844,0.9800259,0.00659607,0.005148032,0.002671019,0.00002013683],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3145759,0.001017453,0.6621363,0.003947741,0.0006241467,0.0004325208,0.005834305,0.007791558,0.003640095],"genre_scores_gemma":[0.743679,0.0005182358,0.2427463,0.000480823,0.0001381135,0.0004273185,0.01002475,0.0001659844,0.001819431],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009809363,"threshold_uncertainty_score":0.01950455,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3206965157579233,"score_gpt":0.4300672946894363,"score_spread":0.109370778931513,"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."}}