{"id":"W1977733667","doi":"10.1093/comjnl/bxt106","title":"A New and Effective Approach to GML Documents Compression","year":2013,"lang":"en","type":"article","venue":"The Computer Journal","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Computer science; XML; Compression ratio; Information retrieval; Data mining; Compression (physics); Encoding (memory); Tree (set theory); Data compression; Pattern matching; Database; Artificial intelligence; 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":[0.0002958691,0.0006652685,0.0004706373,0.00201911,0.0005138299,0.0009962427,0.0009880945,0.0005861678,0.002012931],"category_scores_gemma":[0.001854595,0.0002277819,0.0004304625,0.002691009,0.000546383,0.002036496,0.001108369,0.0007826146,0.001195612],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004095729,"about_ca_system_score_gemma":0.0006549359,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009623619,"about_ca_topic_score_gemma":0.000942422,"domain_scores_codex":[0.9991395,0.00006951039,0.0000886579,0.0001178504,0.0005369884,0.0000475521],"domain_scores_gemma":[0.9992292,0.0001175596,0.00007181503,0.0002279358,0.0003299367,0.00002347727],"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":[0.0002052962,0.00008102994,0.0008068774,0.0003073191,0.00003399346,0.0003898314,0.0002695142,0.009431145,0.09970193,0.01782436,0.006971594,0.863977],"study_design_scores_gemma":[0.000116691,0.0003940654,0.002269431,0.0001177323,0.0001007218,0.003488808,0.0003221897,0.4683184,0.3900036,0.02469157,0.1100792,0.00009766179],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01710503,0.000959467,0.9751061,0.000401015,0.0002320859,0.0001930268,0.0002662627,0.002666356,0.003070538],"genre_scores_gemma":[0.1393954,0.001662449,0.8490546,0.0003594775,0.0002204414,0.0003093282,0.001246076,0.0003100495,0.007442326],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00201911,"threshold_uncertainty_score":0.006733954,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008118702872697147,"score_gpt":0.2224352901072514,"score_spread":0.2143165872345542,"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."}}