{"id":"W4283449126","doi":"10.1145/3530190.3534829","title":"Note: Leveraging Artificial Intelligence to build a Data Catalog and support research on the Sustainable Development Goals","year":2022,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Western University; Université de Montréal","funders":"","keywords":"Metadata; Computer science; Prosperity; Sustainable development; Data science; Sustainability; Big data; World Wide Web; Political science; Data mining","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","sts","open_science","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.04310535,0.0001011482,0.000143695,0.0004215154,0.001694466,0.0008707428,0.00364299,0.00001647218,0.003147225],"category_scores_gemma":[0.004172776,0.0000652572,0.00001490869,0.001575906,0.000113583,0.0003601066,0.01384482,0.0003125314,0.0008002777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001407101,"about_ca_system_score_gemma":0.0003122379,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005862127,"about_ca_topic_score_gemma":0.0004081761,"domain_scores_codex":[0.994629,0.00067999,0.000536606,0.0008725378,0.002738751,0.0005431139],"domain_scores_gemma":[0.9951691,0.002459404,0.00006066797,0.001982072,0.0001977988,0.0001309823],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00006008406,0.0001098666,0.00004502201,0.000006673093,0.000008907177,0.00006442623,0.004306969,0.0002562578,0.00001679601,0.488648,0.2873899,0.2190872],"study_design_scores_gemma":[0.00001650998,0.00008662257,0.0002449318,0.000002535387,0.000001728983,0.000002728834,0.07877689,0.0006132512,0.0009211017,0.06068777,0.8585357,0.0001102314],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1518559,0.00008807528,0.4543262,0.2921194,0.001055353,0.005058679,0.0005592796,0.0001941569,0.0947431],"genre_scores_gemma":[0.9591441,0.000003803318,0.002460681,0.005082105,0.00003881871,0.0001229898,0.00009942961,0.000008455315,0.03303958],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8072883,"threshold_uncertainty_score":0.9999777,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6059346502404944,"score_gpt":0.5244392208282098,"score_spread":0.0814954294122846,"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."}}