{"id":"W2775078427","doi":"10.1177/2053951717745678","title":"Challenges in administrative data linkage for research","year":2017,"lang":"en","type":"article","venue":"Big Data & Society","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":356,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute for Clinical Evaluative Sciences","funders":"Economic and Social Research Council; Wellcome Trust","keywords":"Linkage (software); Record linkage; Computer science; Data quality; Data science; Data collection; Confidentiality; Data mining; Sample (material); Population; Computer security; Engineering; Statistics; Sociology","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"],"domain":"methods","study_design":"not_applicable","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":["metaresearch"],"domain":"methods","study_design":"design_other","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","scholarly_communication","open_science"],"consensus_categories":["metaresearch","open_science"],"category_scores_codex":[0.04328134,0.0001217356,0.0002611787,0.000056614,0.0008379799,0.001330097,0.01724683,0.0001122376,0.00004209236],"category_scores_gemma":[0.01742786,0.00009554727,0.00005861652,0.0001765882,0.0004806106,0.002232797,0.01498031,0.0002988479,0.0002094819],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003457674,"about_ca_system_score_gemma":0.0002224809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002421229,"about_ca_topic_score_gemma":0.007538258,"domain_scores_codex":[0.9956093,0.0003689507,0.0004936482,0.001549033,0.001513541,0.0004654884],"domain_scores_gemma":[0.981907,0.002641019,0.0002371761,0.01489041,0.0002130244,0.0001114057],"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":[0.00001755518,0.000117086,0.0001263642,0.00002794101,0.00002463625,0.000004436773,0.001359644,1.021594e-7,0.000008770851,0.007451819,0.6041026,0.386759],"study_design_scores_gemma":[0.0004770437,0.00005324147,0.01398545,0.00003046324,0.000006787581,4.708933e-7,0.01625402,0.001598255,0.00001628176,0.01735094,0.9500995,0.0001275776],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.02054836,0.01685871,0.05667542,0.4947101,0.008716423,0.0104538,0.1859108,0.0003273233,0.2057991],"genre_scores_gemma":[0.9244187,0.01976064,0.03117278,0.001338234,0.002379258,0.0001834843,0.01221866,0.00004949484,0.008478783],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9038703,"threshold_uncertainty_score":0.9997066,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.9805021151538041,"score_gpt":0.681924870153455,"score_spread":0.2985772450003491,"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."}}