{"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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002195963,0.00005996158,0.0001348706,0.00008640931,0.00004915127,0.00002938416,0.00172675,0.00003219257,0.004583556],"category_scores_gemma":[0.001227561,0.00004081033,0.00005266534,0.0004764517,0.0004658086,0.0002238515,0.0008168292,0.0001073669,0.00008395988],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001812302,"about_ca_system_score_gemma":0.00002595941,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000765829,"about_ca_topic_score_gemma":0.00003020863,"domain_scores_codex":[0.9985893,0.0001327377,0.0004170135,0.0001755616,0.0005594804,0.0001258484],"domain_scores_gemma":[0.9990262,0.0001369032,0.0003721843,0.0002044453,0.0001489071,0.0001113565],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001658532,0.00003118396,0.001990131,0.000001322657,0.00002833853,0.00000213972,0.0002487064,0.00001300637,0.9332578,0.000283412,0.0007178538,0.06326024],"study_design_scores_gemma":[0.001353459,0.001028905,0.4302955,0.000005006499,0.00004959257,0.0001188233,0.002090615,0.001196514,0.4409688,0.001049119,0.121611,0.0002326975],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.929203,0.00003393093,0.04592189,0.0208336,0.001425577,0.00009753682,0.0001603119,0.000007246511,0.00231696],"genre_scores_gemma":[0.9948518,0.00001090891,0.002807309,0.002197474,0.0001113575,0.00000118447,0.000004237594,0.000003168424,0.00001257694],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.492289,"threshold_uncertainty_score":0.9963264,"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."}}