{"id":"W1599200063","doi":"10.1007/s10115-015-0871-2","title":"Partial materialization for online analytical processing over multi-tagged document collections","year":2015,"lang":"en","type":"article","venue":"Knowledge and Information Systems","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"University of Waterloo; Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Computer science; Centroid; Information retrieval; Medoid; Document clustering; Digital library; Data mining; Artificial intelligence; Cluster analysis","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002305905,0.001038611,0.001309015,0.003883582,0.001612597,0.004334277,0.002617714,0.0007460133,0.006841659],"category_scores_gemma":[0.01256918,0.0006911732,0.001260828,0.007802518,0.001356943,0.005637785,0.004732731,0.0008815596,0.004547811],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001135061,"about_ca_system_score_gemma":0.002549262,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002478098,"about_ca_topic_score_gemma":0.003029506,"domain_scores_codex":[0.9978696,0.0004030498,0.0003555913,0.0004327748,0.0007509559,0.000187995],"domain_scores_gemma":[0.9922284,0.002174912,0.0006984078,0.002906693,0.001710452,0.0002809396],"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":[0.001237227,0.0002480588,0.002959429,0.0005707243,0.0001053182,0.0006048274,0.0016574,0.02770588,0.04573967,0.03069949,0.01835776,0.8701143],"study_design_scores_gemma":[0.0003081162,0.001510065,0.005626458,0.0001733516,0.0003013641,0.002697675,0.003767169,0.6165429,0.1749433,0.09046786,0.1033547,0.0003070438],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03123955,0.0004666232,0.954784,0.0003531105,0.00007134584,0.0004474886,0.001376152,0.009003701,0.002257974],"genre_scores_gemma":[0.2157527,0.0005069521,0.7720177,0.0001862498,0.0001407525,0.0007977035,0.004886377,0.0008160646,0.004895428],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006841659,"threshold_uncertainty_score":0.02288765,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04078338301597355,"score_gpt":0.3180951839573361,"score_spread":0.2773118009413625,"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."}}