{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02878123,0.000855526,0.000898416,0.02017461,0.003874743,0.01455043,0.003497188,0.002182705,0.02307513],"category_scores_gemma":[0.07603664,0.001201089,0.001620778,0.02910763,0.003439673,0.02587067,0.01260149,0.004447379,0.01933457],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002347059,"about_ca_system_score_gemma":0.01207164,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01260924,"about_ca_topic_score_gemma":0.02125462,"domain_scores_codex":[0.9857013,0.003900638,0.003004995,0.002152501,0.004693039,0.0005475749],"domain_scores_gemma":[0.8970845,0.03102646,0.004648947,0.04479286,0.0193162,0.003130983],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001638083,0.0001594275,0.01341945,0.001075445,0.0001037556,0.0003714861,0.001950208,0.00230651,0.002967707,0.2063689,0.4137473,0.3573661],"study_design_scores_gemma":[0.0000364575,0.00003263003,0.002958855,0.0004002569,0.00003580964,0.0002828547,0.0006546195,0.004909101,0.001976614,0.07474088,0.9138901,0.00008188323],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008109024,0.002503253,0.7720312,0.05028227,0.003287161,0.003026084,0.06771073,0.0218756,0.07117458],"genre_scores_gemma":[0.02288573,0.001474052,0.9014318,0.002892072,0.0006200051,0.0009001023,0.05823305,0.001848321,0.009714961],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02878123,"threshold_uncertainty_score":0.1522115,"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."}}