{"id":"W4224273000","doi":"10.1016/j.scitotenv.2022.155324","title":"An integrated 3D CNN-GRU deep learning method for short-term prediction of PM2.5 concentration in urban environment","year":2022,"lang":"en","type":"article","venue":"The Science of The Total Environment","topic":"Air Quality Monitoring and Forecasting","field":"Environmental Science","cited_by":141,"is_retracted":false,"has_abstract":false,"ca_institutions":"Institut National de la Recherche Scientifique; Carleton University","funders":"","keywords":"Convolutional neural network; Autoregressive integrated moving average; Deep learning; Air quality index; Computer science; Term (time); Artificial intelligence; Artificial neural network; Time series; Machine learning; Environmental science; Pattern recognition (psychology); Data mining; Meteorology; Geography","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003004674,0.0001414931,0.0001637366,0.0000331054,0.0006672506,0.00001602688,0.0006994646,0.00003096331,0.0003595935],"category_scores_gemma":[0.00005212811,0.0000987575,0.0000772008,0.0002662003,0.0009568852,0.0002257101,0.0005263889,0.0002664955,0.000004335754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007440964,"about_ca_system_score_gemma":0.00001655217,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001708675,"about_ca_topic_score_gemma":9.235779e-7,"domain_scores_codex":[0.9975982,0.0003663799,0.0004173247,0.0003916447,0.0008988892,0.0003275551],"domain_scores_gemma":[0.9991384,0.00009758682,0.0002486253,0.0004485494,0.000003090633,0.00006375997],"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":[0.00003161023,0.0001253985,0.004353375,0.000003684557,0.000004211266,1.426765e-7,0.002736703,0.7456708,0.2344708,0.0000349666,0.000003926137,0.01256439],"study_design_scores_gemma":[0.0004193823,0.0008177937,0.194766,0.00002065712,0.00004566849,0.000009384367,0.003373933,0.7418397,0.05794978,0.0001922484,0.0003629157,0.0002025],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9888108,0.00002921926,0.01003335,0.0001149029,0.0001934105,0.0005870753,0.0000220641,0.00001285156,0.0001963535],"genre_scores_gemma":[0.9948198,0.000009488674,0.004827241,0.000006551771,0.00002666796,0.00008928734,0.000008892496,0.00001155737,0.0002005272],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1904127,"threshold_uncertainty_score":0.513202,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02077895611289331,"score_gpt":0.2551189607768146,"score_spread":0.2343400046639213,"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."}}