{"id":"W3042508634","doi":"10.1007/s00484-020-01968-z","title":"Outlier detection methods to improve the quality of citizen science data","year":2020,"lang":"en","type":"article","venue":"International Journal of Biometeorology","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Outlier; Percentile; Anomaly detection; Computer science; Reliability (semiconductor); Citizen science; Spatial analysis; Autocorrelation; Statistics; Cluster analysis; Data mining; Artificial intelligence; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01673985,0.001137926,0.00174184,0.004692866,0.00100569,0.003321024,0.002975661,0.001931385,0.002226596],"category_scores_gemma":[0.08240101,0.0005706128,0.001245744,0.005902552,0.0008513672,0.003153763,0.003233342,0.002589841,0.001344507],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007143919,"about_ca_system_score_gemma":0.00215335,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004771154,"about_ca_topic_score_gemma":0.005349763,"domain_scores_codex":[0.9883249,0.003646139,0.001352926,0.002072645,0.004019307,0.0005841193],"domain_scores_gemma":[0.9278958,0.04008206,0.005034486,0.01191863,0.01417331,0.0008957313],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001175291,0.0007339825,0.1538836,0.0007758213,0.001009051,0.0005014877,0.0008163838,0.09028441,0.02197772,0.01706554,0.03033889,0.6814378],"study_design_scores_gemma":[0.0001183379,0.0001678447,0.01925782,0.00008826392,0.0001311376,0.0003659575,0.0003428221,0.9247953,0.01419439,0.02659844,0.01385881,0.00008099336],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0497872,0.0005657417,0.9410542,0.0007607476,0.0003075054,0.000156567,0.002393064,0.004146884,0.0008279921],"genre_scores_gemma":[0.4685071,0.0004198473,0.5152379,0.0004136991,0.0004073333,0.000305938,0.01241492,0.0007124736,0.001580708],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9832602,"threshold_uncertainty_score":0.08852988,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1148161789629101,"score_gpt":0.4244952406370947,"score_spread":0.3096790616741845,"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."}}