{"id":"W2141277572","doi":"10.1109/tkde.2013.53","title":"Extended Subtree: A New Similarity Function for Tree Structured Data","year":2013,"lang":"en","type":"article","venue":"IEEE Transactions on Knowledge and Data Engineering","topic":"Advanced Clustering Algorithms Research","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Similarity (geometry); Edit distance; Cluster analysis; Data mining; Tree (set theory); Function (biology); Theoretical computer science; Artificial intelligence; Mathematics","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":[0.0001817815,0.000198937,0.0001769794,0.0001759333,0.0001674748,0.0002178093,0.001627191,0.00008475543,0.00002101199],"category_scores_gemma":[0.00002963129,0.0001957729,0.00002769348,0.000325157,0.00002076066,0.002130516,0.00008591426,0.0002695416,0.00002582856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004077655,"about_ca_system_score_gemma":0.00008051616,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000488581,"about_ca_topic_score_gemma":0.0001080283,"domain_scores_codex":[0.9984848,0.00001777156,0.0001994576,0.0007802632,0.0001621853,0.0003554524],"domain_scores_gemma":[0.997189,0.0002104163,0.00002677364,0.002279489,0.00007526279,0.0002190369],"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":[0.00001999081,0.00007365351,0.000001096664,0.00008858158,0.00006453335,0.000001572279,0.0001119688,0.006567033,0.004280792,0.0001393087,0.002770263,0.9858812],"study_design_scores_gemma":[0.0006282991,0.0001022651,0.0004396339,0.00002665284,0.00002034684,0.0000122096,0.000009937925,0.9872869,0.001693562,0.0003292407,0.009231338,0.0002195811],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0002911379,0.0002965762,0.9975629,0.0001752486,0.0007290294,0.0004063013,0.0002319387,0.0002726035,0.00003425726],"genre_scores_gemma":[0.4898209,0.0001622713,0.5079281,0.0000583659,0.0003573079,0.0001318782,0.000281712,0.0000727162,0.001186718],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9856616,"threshold_uncertainty_score":0.7983387,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05833427377449112,"score_gpt":0.3147285319055442,"score_spread":0.256394258131053,"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."}}