{"id":"W2080801059","doi":"10.1109/wi.2004.42","title":"Clustering Web Surfers with Probabilistic Models in a Real Application","year":2004,"lang":"en","type":"article","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Open Text (Canada); York University","funders":"","keywords":"Cluster analysis; Computer science; Data mining; Probabilistic logic; Web mining; Machine learning; Web page; Artificial intelligence; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000119458,0.00006149348,0.00005872284,0.00006498715,0.00002278415,0.00009226456,0.0003881446,0.0000128594,0.000001017317],"category_scores_gemma":[0.000002372128,0.00004827846,0.000007921234,0.000302726,0.00001421478,0.0007680543,0.0001583599,0.0000358662,0.00001488938],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005084446,"about_ca_system_score_gemma":0.0000296631,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003577935,"about_ca_topic_score_gemma":0.001039125,"domain_scores_codex":[0.9993978,0.000006481378,0.00009101239,0.0002462052,0.0001204994,0.0001380552],"domain_scores_gemma":[0.9996035,0.000006748583,0.00002087625,0.0003216075,0.00002017227,0.0000271533],"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":[0.00001009116,0.0001260552,0.000381543,0.00003984035,0.000007446276,0.00001545793,0.0004479372,0.397063,0.00007001998,0.55792,0.00006029546,0.04385838],"study_design_scores_gemma":[0.0004057593,0.00003251712,0.000477321,0.00001247751,0.000001288828,0.000001043904,0.00001706507,0.9854301,0.00001538029,0.01341115,0.0001054414,0.00009049271],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008671928,0.00000186221,0.9691337,0.0003193429,0.00002550662,0.0002342891,5.229838e-7,0.0001342337,0.02147866],"genre_scores_gemma":[0.8526098,0.000006892827,0.1470925,0.00007402735,0.00001360024,0.0000404367,0.00000509037,0.000003951122,0.0001537045],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8439379,"threshold_uncertainty_score":0.1968739,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01942104449193428,"score_gpt":0.2298075427378946,"score_spread":0.2103864982459603,"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."}}