{"id":"W4385571965","doi":"10.18653/v1/2023.findings-acl.604","title":"AnaMeta: A Table Understanding Dataset of Field Metadata Knowledge Shared by Multi-dimensional Data Analysis Tasks","year":2023,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Metadata; Computer science; Field (mathematics); Metadata repository; Table (database); Information retrieval; Dimension (graph theory); Data element; Data type; Data mining; World Wide Web","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":[0.001586692,0.001265521,0.0008255395,0.004331646,0.001157716,0.002027831,0.002445639,0.001629257,0.003733212],"category_scores_gemma":[0.01098699,0.00046595,0.00163228,0.005191008,0.0006740148,0.002784508,0.002029513,0.001631261,0.003116934],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001658423,"about_ca_system_score_gemma":0.002980061,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0238582,"about_ca_topic_score_gemma":0.05336275,"domain_scores_codex":[0.9979171,0.0003594182,0.0003685009,0.0006286647,0.000602273,0.0001239639],"domain_scores_gemma":[0.9927598,0.002266963,0.0008136519,0.002044902,0.00155795,0.0005567262],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00157514,0.00146545,0.08483682,0.00420949,0.0007922995,0.001125343,0.001191758,0.01859158,0.013828,0.009785661,0.7028996,0.1596989],"study_design_scores_gemma":[0.001088675,0.0007203225,0.180031,0.0007487769,0.0003660042,0.001954596,0.002930675,0.1339226,0.01659161,0.02888223,0.6323398,0.0004238125],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.08839113,0.001634736,0.02117739,0.00130616,0.000290164,0.0005787201,0.8659722,0.01473393,0.005915607],"genre_scores_gemma":[0.07394949,0.000396975,0.05785045,0.0003079695,0.00007348254,0.0006027638,0.8651701,0.0003084528,0.00134037],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.0238582,"threshold_uncertainty_score":0.04743862,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6644636814897413,"score_gpt":0.5129623906594967,"score_spread":0.1515012908302447,"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."}}