{"id":"W4241938281","doi":"10.1145/376284.375672","title":"Data bubbles","year":2001,"lang":"en","type":"article","venue":"ACM SIGMOD Record","topic":"Advanced Clustering Algorithms Research","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Cluster analysis; Computer science; Data mining; CURE data clustering algorithm; Data stream clustering; Canopy clustering algorithm; Hierarchical clustering; Correlation clustering; Set (abstract data type); Data set; Data compression; Key (lock); Constrained clustering; Determining the number of clusters in a data set; Fuzzy clustering; Algorithm; Artificial intelligence","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":[0.004386181,0.0007071546,0.0009842973,0.002273777,0.001726156,0.003841629,0.003085714,0.001957915,0.02465323],"category_scores_gemma":[0.02955338,0.0005852928,0.001068008,0.002997718,0.002020341,0.01169514,0.006867439,0.00255946,0.007902516],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001306932,"about_ca_system_score_gemma":0.001255591,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001573575,"about_ca_topic_score_gemma":0.001147603,"domain_scores_codex":[0.9950486,0.001115173,0.0003746202,0.001095569,0.002111988,0.0002540601],"domain_scores_gemma":[0.9850007,0.00683797,0.0007109692,0.004603016,0.002271746,0.0005755757],"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":[0.0005263084,0.0001245578,0.003084861,0.0008460607,0.00008355593,0.0006828432,0.001362703,0.008804914,0.005359281,0.5522953,0.07293121,0.3538984],"study_design_scores_gemma":[0.00007453088,0.0001745935,0.0008361073,0.0002262908,0.0000469375,0.001445506,0.0007936704,0.03794724,0.007958464,0.3173856,0.6330255,0.00008550959],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01669408,0.00478333,0.9249112,0.008200641,0.002105885,0.0007143421,0.003767193,0.004027247,0.03479612],"genre_scores_gemma":[0.1894106,0.004144481,0.7515159,0.00576521,0.001150831,0.001356493,0.009064232,0.001910399,0.03568188],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02465323,"threshold_uncertainty_score":0.08247328,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.123803096386827,"score_gpt":0.366511782884858,"score_spread":0.242708686498031,"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."}}