{"id":"W4242624664","doi":"10.31219/osf.io/ndp8m","title":"Crowdsourcing for machine learning in public health surveillance: lessons learned from Amazon Mechanical Turk","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Mobile Crowdsensing and Crowdsourcing","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Agency of Canada; University of Calgary","funders":"Natural Sciences and Engineering Research Council of Canada; Cumming School of Medicine, University of Calgary; Public Health Agency; Public Health Agency of Canada","keywords":"Crowdsourcing; Computer science; Artificial intelligence; Machine learning; Convolutional neural network; Data science; Inference; Quality (philosophy); Citizen science; Context (archaeology); Deep learning; Leverage (statistics); Ground truth; Natural language processing; World Wide Web","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":["metaepi_narrow","scholarly_communication","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.00413577,0.00066816,0.001397755,0.0004179124,0.0004843991,0.002132047,0.001663246,0.0006141304,0.00005203277],"category_scores_gemma":[0.001206909,0.0006918807,0.0004570821,0.0006629028,0.00005373722,0.0002842524,0.002952343,0.002490451,0.00001019594],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005124858,"about_ca_system_score_gemma":0.001436164,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006197104,"about_ca_topic_score_gemma":0.004881255,"domain_scores_codex":[0.9932761,0.001183592,0.00113764,0.002404353,0.00063726,0.001361075],"domain_scores_gemma":[0.9957155,0.00116013,0.0006303333,0.001773495,0.0002725523,0.0004479755],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007344288,0.0007845582,0.01082422,0.001049306,0.0004331726,0.0002080548,0.01215659,0.07787348,0.004503013,0.02431908,0.001467432,0.8663077],"study_design_scores_gemma":[0.001429682,0.0001402818,0.003140216,0.0007077438,0.00001003899,0.0000221834,0.0007544829,0.9755771,0.00096907,0.006497051,0.009496772,0.001255433],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07815505,0.001674993,0.8804706,0.03634432,0.001403714,0.0006561669,0.00003575269,0.0009120362,0.0003473545],"genre_scores_gemma":[0.9173253,0.0003415374,0.07997593,0.0009776164,0.0003019368,0.0001005524,0.0004413474,0.00008595071,0.0004498644],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8977036,"threshold_uncertainty_score":0.9998108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1054447398296236,"score_gpt":0.3199346824422148,"score_spread":0.2144899426125912,"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."}}