{"id":"W4285801284","doi":"10.3390/atmos13071144","title":"An Improved Air Quality Index Machine Learning-Based Forecasting with Multivariate Data Imputation Approach","year":2022,"lang":"en","type":"article","venue":"Atmosphere","topic":"Air Quality Monitoring and Forecasting","field":"Environmental Science","cited_by":69,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Imputation (statistics); Missing data; Air quality index; Computer science; Artificial neural network; Random forest; Multivariate statistics; Predictive modelling; Data mining; Machine learning; Feature selection; Artificial intelligence; Meteorology","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":[0.002233785,0.0007005315,0.001862178,0.0009750229,0.0004869305,0.0007323725,0.002204224,0.0008319007,0.001373876],"category_scores_gemma":[0.003856845,0.000445632,0.00196009,0.001956617,0.0002155199,0.001018124,0.0007836508,0.00176582,0.0004608452],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005521957,"about_ca_system_score_gemma":0.001344507,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01432697,"about_ca_topic_score_gemma":0.008730207,"domain_scores_codex":[0.999182,0.00024594,0.00007072831,0.0001989659,0.000200409,0.0001018279],"domain_scores_gemma":[0.9985908,0.0006121042,0.0001460252,0.0001692683,0.0004367475,0.00004504749],"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":[0.0001063271,0.000122308,0.005824449,0.00008321469,0.0002044333,0.0001142963,0.00007058986,0.8310946,0.001222813,0.003766168,0.002060476,0.1553303],"study_design_scores_gemma":[0.000003036476,0.00000910088,0.0003002446,0.000003200393,0.000009457247,0.000006747307,0.000002382817,0.9987155,0.0001543313,0.0005390096,0.0002529866,0.000004055357],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01745386,0.0003339068,0.9799571,0.0001989162,0.00008861705,0.00003050135,0.0002865642,0.0006791437,0.0009713803],"genre_scores_gemma":[0.5676749,0.0007323355,0.4242477,0.0002536641,0.0003698031,0.0002528501,0.002415053,0.0001432936,0.003910381],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01432697,"threshold_uncertainty_score":0.02848715,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04977443104511185,"score_gpt":0.2850346989494251,"score_spread":0.2352602679043132,"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."}}