{"id":"W2117544284","doi":"10.1145/375663.375672","title":"Data bubbles","year":2001,"lang":"en","type":"article","venue":"","topic":"Advanced Clustering Algorithms Research","field":"Computer Science","cited_by":79,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Cluster analysis; Computer science; CURE data clustering algorithm; Canopy clustering algorithm; Data mining; Data stream clustering; Correlation clustering; Hierarchical clustering; Data compression; Data set; Set (abstract data type); Fuzzy clustering; Constrained clustering; Determining the number of clusters in a data set; 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":[0.004271393,0.000717883,0.001015422,0.002230354,0.001585356,0.003580371,0.002917317,0.001906889,0.02160664],"category_scores_gemma":[0.02874354,0.0005578107,0.001041123,0.002926084,0.001866302,0.01051364,0.006357067,0.002424677,0.007120849],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001190537,"about_ca_system_score_gemma":0.001185502,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001490794,"about_ca_topic_score_gemma":0.001056572,"domain_scores_codex":[0.9951416,0.001068352,0.0003799366,0.001106916,0.00206573,0.0002374713],"domain_scores_gemma":[0.9858145,0.006481198,0.0006928246,0.004297899,0.002199639,0.000514006],"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":[0.0005966222,0.00013087,0.003386099,0.0009590957,0.00009543943,0.0007352084,0.001355408,0.01140886,0.006778263,0.5116902,0.06766343,0.3952005],"study_design_scores_gemma":[0.00008394467,0.0002045245,0.001039186,0.0002506467,0.00005381997,0.001624107,0.0008639693,0.05318266,0.01080382,0.3285155,0.6032785,0.00009935022],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01599324,0.004075348,0.9381399,0.006197358,0.001712561,0.0006598126,0.0034109,0.0034226,0.02638833],"genre_scores_gemma":[0.1833804,0.003671493,0.767874,0.004676725,0.0009755068,0.001301112,0.008454241,0.001591,0.02807548],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02160664,"threshold_uncertainty_score":0.07228148,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1292531724393648,"score_gpt":0.3804818146787029,"score_spread":0.2512286422393381,"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."}}