{"id":"W2061022164","doi":"10.1145/1451940.1451960","title":"A bottom-up approach for XML documents classification","year":2008,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"","keywords":"Computer science; Document Structure Description; Information retrieval; XML; SGML; XML Schema Editor; XML validation; Efficient XML Interchange; Document type definition; XML Schema (W3C); XML database; Schema (genetic algorithms); 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":[0.001870154,0.001607122,0.001770345,0.008751831,0.002305281,0.00338642,0.003140344,0.002584795,0.003437674],"category_scores_gemma":[0.004502883,0.0005694999,0.002709446,0.006514924,0.001062245,0.002752258,0.0023171,0.002442996,0.004292121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009587548,"about_ca_system_score_gemma":0.00164908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003917887,"about_ca_topic_score_gemma":0.004335927,"domain_scores_codex":[0.9947777,0.0007302437,0.0004990746,0.0009259802,0.002802921,0.0002640209],"domain_scores_gemma":[0.9969727,0.0008456589,0.000192152,0.0006528539,0.001202658,0.00013393],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009277344,0.0002431755,0.001992316,0.000292232,0.0001106007,0.0001809978,0.0002195073,0.00626734,0.01103687,0.006853147,0.01241275,0.9602983],"study_design_scores_gemma":[0.00007779914,0.0003954153,0.004779167,0.0002918588,0.0003643857,0.001930235,0.0004893283,0.7658583,0.05241698,0.07219787,0.1009411,0.000257618],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002682755,0.0006959802,0.9918244,0.0002180879,0.0001145686,0.000370503,0.0004261867,0.002518003,0.00114944],"genre_scores_gemma":[0.02644487,0.0004399244,0.9674841,0.0002393233,0.0001663393,0.0003236503,0.002026809,0.0001268691,0.002748125],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008751831,"threshold_uncertainty_score":0.01150012,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0649444665900192,"score_gpt":0.2806387728962791,"score_spread":0.2156943063062599,"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."}}