{"id":"W4312095289","doi":"10.1016/j.envint.2022.107691","title":"Multi-step forecast of PM2.5 and PM10 concentrations using convolutional neural network integrated with spatial–temporal attention and residual learning","year":2022,"lang":"en","type":"article","venue":"Environment International","topic":"Air Quality Monitoring and Forecasting","field":"Environmental Science","cited_by":91,"is_retracted":false,"has_abstract":true,"ca_institutions":"Lakes Environmental (Canada); University of Waterloo","funders":"Xuzhou Science and Technology Bureau; Jiangsu Provincial Department of Human Resources and Social Security; Jiangsu Provincial Department of Education; National Natural Science Foundation of China","keywords":"Residual; Computer science; Artificial neural network; Convolutional neural network; Baseline (sea); Meteorology; Mean squared error; Environmental science; Generalization; Stability (learning theory); Data mining; Statistics; Machine learning; Mathematics; Algorithm; 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":[0.0002862501,0.0001184677,0.0001090285,0.00002840246,0.0003921922,0.00002342008,0.00008996103,0.00002706546,0.0008738378],"category_scores_gemma":[0.00001596749,0.0001174288,0.00002318734,0.00005723092,0.0002814422,0.0001541815,0.0002761271,0.0002211671,0.000003299066],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001810602,"about_ca_system_score_gemma":0.000009314094,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006516103,"about_ca_topic_score_gemma":0.00004507616,"domain_scores_codex":[0.998771,0.0001245869,0.0002454723,0.0002658604,0.0004253242,0.0001678251],"domain_scores_gemma":[0.9996194,0.00005432426,0.0001990769,0.00006483708,0.000006066451,0.00005625388],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000617878,0.00005805749,0.704402,0.00000243247,0.00002744716,0.000003394476,0.0002417642,0.2904786,0.002200313,0.00004952019,0.00005893721,0.00241577],"study_design_scores_gemma":[0.0007644449,0.0001902778,0.3938649,0.00001754344,0.00002298438,0.00005244935,0.0008983757,0.6006858,0.00005552152,0.00002366149,0.00326575,0.0001583117],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9716207,0.00004355061,0.02772965,0.0001533698,0.0001655691,0.0001370069,0.00005655802,0.00001651924,0.00007707068],"genre_scores_gemma":[0.9844736,0.000008224026,0.0147815,0.00002301323,0.00008425453,0.00001615631,0.000230502,0.0000122874,0.0003703885],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3105371,"threshold_uncertainty_score":0.9567911,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02751396669220346,"score_gpt":0.2379162000873369,"score_spread":0.2104022333951334,"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."}}