{"id":"W2109010760","doi":"10.1109/ssdbm.2006.6","title":"A Disc-based Approach to Data Summarization and Privacy Preservation","year":2006,"lang":"en","type":"article","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Automatic summarization; Computer science; Data mining; Heuristic; Constraint (computer-aided design); Set (abstract data type); Cluster analysis; Exploit; Algorithm; Theoretical computer science; Information retrieval; 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":[0.006407493,0.0009806504,0.001764608,0.004388091,0.001387175,0.003578866,0.002811514,0.0018791,0.002285022],"category_scores_gemma":[0.02809863,0.0005059617,0.001295608,0.005901217,0.002245536,0.006400976,0.00394297,0.002453095,0.001111679],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001495131,"about_ca_system_score_gemma":0.001297213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006220319,"about_ca_topic_score_gemma":0.000483811,"domain_scores_codex":[0.9921039,0.002870746,0.0009839472,0.001159486,0.002609525,0.0002722576],"domain_scores_gemma":[0.9799373,0.007269791,0.001963593,0.007265701,0.003183763,0.0003799362],"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":[0.0006650124,0.0001780428,0.002126597,0.0008275968,0.0001740944,0.0002562599,0.001012507,0.1182398,0.01426956,0.2549112,0.0137083,0.593631],"study_design_scores_gemma":[0.0001234961,0.0008524918,0.001212346,0.0002158708,0.000179206,0.001322786,0.00056354,0.5310396,0.0428775,0.3496528,0.07184046,0.0001199745],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003744381,0.000536358,0.9929702,0.0005181138,0.00006491829,0.0001260813,0.0002566139,0.00044655,0.001336762],"genre_scores_gemma":[0.1390048,0.0008999296,0.8550583,0.0003978024,0.0003589111,0.0004321992,0.001262142,0.0001686222,0.002417315],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006407493,"threshold_uncertainty_score":0.03388649,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04332015707856461,"score_gpt":0.2489406957837209,"score_spread":0.2056205387051563,"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."}}