{"id":"W4387097696","doi":"10.1038/s41612-023-00475-3","title":"Improving air quality assessment using physics-inspired deep graph learning","year":2023,"lang":"en","type":"article","venue":"npj Climate and Atmospheric Science","topic":"Air Quality Monitoring and Forecasting","field":"Environmental Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Key Research and Development Program of China; Chinese Academy of Sciences; Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Science; State Key Laboratory of Resources and Environmental Information System; National Natural Science Foundation of China; Nvidia","keywords":"Extrapolation; Air quality index; Artificial neural network; Pollutant; Graph; Computer science; Air pollution; Reliability (semiconductor); Air pollutants; Artificial intelligence; Machine learning; Environmental science; Data mining; Meteorology; Mathematics; Statistics; Geography; Physics; Theoretical computer science","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.002220242,0.0001805017,0.00019794,0.000007511563,0.001627204,0.0001451349,0.0002960324,0.00004663364,0.0000485149],"category_scores_gemma":[0.0001378759,0.0001669734,0.00005606444,0.001790358,0.0007522177,0.000661312,0.0006384379,0.000213884,0.00006542405],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000194859,"about_ca_system_score_gemma":0.00003494909,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008009349,"about_ca_topic_score_gemma":0.000008035076,"domain_scores_codex":[0.9975419,0.00008456847,0.0002985527,0.000634012,0.0006738054,0.0007671303],"domain_scores_gemma":[0.9992447,0.0001090807,0.0002003256,0.0002412511,0.00001998233,0.0001846669],"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":[0.000007046286,0.00004177992,0.5816326,0.0000532757,0.000004885349,0.00000924059,0.001375558,0.05499421,0.1494588,0.000259084,0.00000723323,0.2121563],"study_design_scores_gemma":[0.0002077371,0.00006874208,0.415825,0.00003190736,0.00001164332,0.000004759183,0.002177872,0.5797254,0.001099628,0.000409421,0.0001109764,0.0003268466],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9895244,0.00001731104,0.007778823,0.00004696767,0.0002747855,0.0001047473,0.000001286625,0.0002677779,0.001983928],"genre_scores_gemma":[0.9761413,0.00003795196,0.02356375,0.00006146821,0.00008408525,0.000007842819,0.000001788328,0.00001610565,0.00008570149],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5247312,"threshold_uncertainty_score":0.9996725,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04431416067510092,"score_gpt":0.3195652019720248,"score_spread":0.2752510412969238,"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."}}