{"id":"W3011864725","doi":"","title":"DescribeX: A Framework for Exploring and Querying XML Web Collections","year":2008,"lang":"en","type":"preprint","venue":"TSpace","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"XPath; Computer science; Automatic summarization; XML; Information retrieval; Path expression; Scalability; XML database; Streaming XML; Database; Key (lock); Search engine indexing; World Wide Web","routes":{"ca_aff":true,"ca_fund":true,"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":[0.008835471,0.002160657,0.002786792,0.004776319,0.002258785,0.009951605,0.009289435,0.003695179,0.01067326],"category_scores_gemma":[0.01236113,0.003874478,0.005273892,0.005268756,0.002932417,0.01382298,0.01073918,0.00466115,0.005025206],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002556184,"about_ca_system_score_gemma":0.003013051,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01387744,"about_ca_topic_score_gemma":0.01691866,"domain_scores_codex":[0.9937851,0.001552021,0.001066438,0.0008643335,0.002393627,0.0003384285],"domain_scores_gemma":[0.9958474,0.001840231,0.0002805169,0.0012839,0.0004613656,0.0002866078],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008055465,0.0002875851,0.005613978,0.004264878,0.0007271938,0.00235201,0.003871155,0.05702611,0.01112889,0.458245,0.1350347,0.3206429],"study_design_scores_gemma":[0.0002693871,0.000121431,0.0008171994,0.0007117197,0.000215449,0.001642564,0.0008527797,0.2074191,0.0111747,0.1888098,0.5876501,0.0003159311],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0009027775,0.0005592706,0.9634731,0.0003287596,0.00005636742,0.0004288435,0.002024828,0.03032913,0.001896948],"genre_scores_gemma":[0.02291489,0.002147775,0.9450908,0.0006835528,0.0001109295,0.001601121,0.01379484,0.007122542,0.006533547],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01387744,"threshold_uncertainty_score":0.04672694,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1304925359317852,"score_gpt":0.345237154838614,"score_spread":0.2147446189068288,"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."}}