{"id":"W2528898284","doi":"","title":"Record linkage for web data","year":2012,"lang":"en","type":"dissertation","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Record linkage; Linkage (software); Linked data; Relational database; Task (project management); Information retrieval; Process (computing); Set (abstract data type); Matching (statistics); Data mining; Database; World Wide Web; Semantic Web; Data science; Engineering; Programming language","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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.006157569,0.0001678056,0.0003234646,0.000220816,0.0001123042,0.0004051746,0.00360867,0.0001814281,0.005351386],"category_scores_gemma":[0.003012062,0.0001194858,0.00009500762,0.000238979,0.00001442182,0.000837237,0.0004281212,0.000118606,0.003095288],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001253908,"about_ca_system_score_gemma":0.00008475109,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001052858,"about_ca_topic_score_gemma":0.006788195,"domain_scores_codex":[0.9972281,0.00009066706,0.0007075516,0.0007495094,0.0009659645,0.0002582038],"domain_scores_gemma":[0.995344,0.001061922,0.0003646305,0.002947559,0.0001862612,0.00009565866],"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":[0.00002287929,0.0000307048,0.00001991169,0.00003323094,0.00001740174,2.557884e-7,0.00006958194,3.793615e-8,0.00001399246,0.008610937,0.6646942,0.3264868],"study_design_scores_gemma":[0.000136176,0.00001849675,0.001461138,0.00001613774,0.00005006639,9.982813e-8,0.002053031,0.0002227541,0.00002043038,0.01176174,0.9840824,0.0001775511],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.004261981,0.0006534621,0.02926572,0.00182861,0.01723761,0.00209588,0.006984711,0.000204371,0.9374676],"genre_scores_gemma":[0.005819385,0.00009009452,0.01235241,0.0007892202,0.0007343859,0.00007186036,0.04699935,0.00002924088,0.9331141],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.3263092,"threshold_uncertainty_score":0.9976809,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5393046271116645,"score_gpt":0.5347174130476245,"score_spread":0.004587214064039946,"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."}}