{"id":"W4410021886","doi":"10.1515/em-2024-0026","title":"Bot invasion: protecting the integrity of online surveys against spamming","year":2025,"lang":"en","type":"article","venue":"Epidemiologic Methods","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vancouver Coastal Health","funders":"University of Sydney; Western Sydney Local Health District","keywords":"Spamming; Internet privacy; Computer security; Business; Computer science; World Wide Web; The Internet","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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.07137387,0.0001500527,0.0004169904,0.0001078068,0.0003601182,0.00003453224,0.001090631,0.0001677515,0.000006308018],"category_scores_gemma":[0.05184156,0.00008899687,0.0001434917,0.0009205762,0.0001006883,0.0001427924,0.0004683196,0.001114949,0.000002380658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004668574,"about_ca_system_score_gemma":0.00008783924,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003910995,"about_ca_topic_score_gemma":0.00004665696,"domain_scores_codex":[0.9761404,0.02239689,0.0006299682,0.0004433585,0.0001195165,0.0002698671],"domain_scores_gemma":[0.9825256,0.01618505,0.0003468426,0.0007816447,0.0001212732,0.00003957759],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000004613663,0.00003461785,0.00659481,0.00003598681,0.00002471041,0.000001208383,0.0002398202,0.0004324896,0.005892694,0.009415735,0.000364739,0.9769586],"study_design_scores_gemma":[0.0005413079,0.0003423661,0.2535633,0.0004942895,0.00003799802,0.00002007751,0.0006162566,0.3103357,0.1068934,0.3113461,0.01522006,0.0005893006],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02453976,0.0004334895,0.9672843,0.004574879,0.001171613,0.0002568816,0.000001101782,0.000147082,0.00159085],"genre_scores_gemma":[0.2503702,0.00004658141,0.7481258,0.001199602,0.00009531358,0.00002104579,0.0000016411,0.000003432586,0.0001364056],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9763693,"threshold_uncertainty_score":0.956216,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1642048597487808,"score_gpt":0.433027778464514,"score_spread":0.2688229187157332,"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."}}