{"id":"W4392557674","doi":"10.1002/leap.1600","title":"Using automated analysis of the bibliography to detect potential research integrity issues","year":2024,"lang":"en","type":"article","venue":"Learned Publishing","topic":"Academic integrity and plagiarism","field":"Social Sciences","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Honeywell (Canada)","funders":"","keywords":"Workflow; Computer science; Data science; Key (lock); Research integrity; Bibliography; Data integrity; Information retrieval; Database; Library science; Computer security; Engineering ethics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["metaresearch","research_integrity","bibliometrics"],"domain":"evaluation","study_design":"simulation_or_modeling","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":["metaresearch","research_integrity"],"domain":"methods","study_design":"observational","genre":"methods","about_ca_system":false,"about_ca_topic":false,"confidence":"medium","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["bibliometrics","scholarly_communication","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0150416,0.0001057623,0.0002387671,0.008951196,0.0009499051,0.003268326,0.001227143,0.0007767689,0.0003520729],"category_scores_gemma":[0.005505029,0.00007887266,0.0003494896,0.06600222,0.0003710226,0.001864921,0.0003025253,0.004231473,0.00001598335],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000986333,"about_ca_system_score_gemma":0.0003220768,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.1047427,"about_ca_topic_score_gemma":0.005424292,"domain_scores_codex":[0.9951464,0.001997846,0.0003101928,0.000379508,0.001607566,0.0005584781],"domain_scores_gemma":[0.9981339,0.0007403244,0.00006501012,0.000297721,0.0006130285,0.0001500314],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001188673,0.00009758708,0.03271523,0.0002081282,0.003965731,0.00005617622,0.1538842,0.01139674,0.01795819,0.1452769,0.5891402,0.04518206],"study_design_scores_gemma":[0.0002498622,0.00009172768,0.03313873,0.001013425,0.001747528,0.000005673291,0.06457151,0.3124535,0.004083143,0.05647495,0.5252427,0.0009272388],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8680468,0.002270488,0.004938408,0.08255488,0.002249698,0.0005962524,0.00008289744,0.001115012,0.0381456],"genre_scores_gemma":[0.9969181,0.0002080049,0.001646798,0.000243219,0.0004041688,0.000004651896,0.000006250503,0.0000124614,0.0005563453],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3010567,"threshold_uncertainty_score":0.9980658,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.144943481836799,"score_gpt":0.4536236169249125,"score_spread":0.3086801350881135,"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."}}