{"id":"W2137580272","doi":"10.1007/s11280-005-1544-y","title":"Studying the XML Web: Gathering Statistics from an XML Sample","year":2005,"lang":"en","type":"article","venue":"World Wide Web","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":76,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Institut national de recherche en informatique et en automatique (INRIA)","keywords":"Computer science; XML validation; Efficient XML Interchange; XML Schema Editor; XML Base; Streaming XML; Document Structure Description; XML Signature; XML Encryption; Information retrieval; World Wide Web; XML database; cXML; XML; Database","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":[],"consensus_categories":[],"category_scores_codex":[0.0002765235,0.0001642383,0.0001779912,0.00005789367,0.0003241576,0.0001415888,0.0006087224,0.00001831619,0.0000877421],"category_scores_gemma":[0.0001008999,0.000120982,0.00003266999,0.0002767713,0.00005746041,0.0007541244,0.0003049779,0.0001661396,0.0001024962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005061784,"about_ca_system_score_gemma":0.00006895511,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004973633,"about_ca_topic_score_gemma":0.01127283,"domain_scores_codex":[0.9986674,0.0000923418,0.0002769695,0.0003747762,0.0002740943,0.0003144622],"domain_scores_gemma":[0.9982024,0.0006047207,0.0001125691,0.0009379047,0.00004422102,0.00009817519],"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":[0.00002357697,0.000191882,0.01674887,0.00002934468,0.00009911246,0.00006017298,0.007051683,0.004902028,0.004503118,0.6974509,0.02345475,0.2454845],"study_design_scores_gemma":[0.0002498568,0.00002479608,0.00255271,0.00004118465,0.000007442942,0.00000279766,0.0002292666,0.07355724,0.0003069788,0.002143076,0.920643,0.0002416896],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01930918,0.0003404383,0.9772409,0.001095631,0.0004664154,0.0001772598,0.000347041,0.000224432,0.0007987552],"genre_scores_gemma":[0.4280125,0.00002355577,0.5693811,0.00120318,0.000507595,0.00002636696,0.00003662204,0.00002327834,0.0007858186],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8971882,"threshold_uncertainty_score":0.6290505,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02279868175895083,"score_gpt":0.2698407601171572,"score_spread":0.2470420783582064,"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."}}