{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002086513,0.0006251882,0.001111479,0.004165016,0.0006434213,0.00164511,0.001660442,0.001245232,0.001469022],"category_scores_gemma":[0.009092638,0.0002814686,0.001180202,0.0056081,0.0006539028,0.005657437,0.002603821,0.001161706,0.001104382],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007460965,"about_ca_system_score_gemma":0.001080134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001039181,"about_ca_topic_score_gemma":0.001489607,"domain_scores_codex":[0.9965545,0.0006672787,0.0004495698,0.0004965391,0.001703395,0.0001287311],"domain_scores_gemma":[0.9962134,0.001222057,0.0004682479,0.0008346188,0.001045452,0.0002162889],"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":[0.0003540624,0.000200775,0.005824598,0.0005695277,0.0002183061,0.0003162291,0.0008288116,0.04002279,0.03328612,0.05511758,0.009282807,0.8539785],"study_design_scores_gemma":[0.00009091879,0.0009738301,0.006187787,0.0001789421,0.0001482496,0.002783946,0.0005954216,0.7546619,0.03248584,0.1347498,0.06693655,0.0002068368],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01855232,0.0007254852,0.9778242,0.0001398301,0.00009570501,0.00009858533,0.0004601135,0.001252562,0.0008512388],"genre_scores_gemma":[0.1731043,0.0006824738,0.821148,0.0001966783,0.0001502186,0.000292396,0.002296607,0.0004138959,0.001715481],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004165016,"threshold_uncertainty_score":0.01103473,"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."}}