{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005321816,0.0007489092,0.001071257,0.002386061,0.0007508314,0.001359126,0.0019047,0.002076302,0.0006182816],"category_scores_gemma":[0.01314398,0.000721005,0.0007470417,0.002564574,0.0006419338,0.002303761,0.0008570302,0.001103098,0.0005155499],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008208816,"about_ca_system_score_gemma":0.0005709867,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008605139,"about_ca_topic_score_gemma":0.0102337,"domain_scores_codex":[0.9976767,0.001146737,0.0001301871,0.0005012469,0.0004272115,0.0001179085],"domain_scores_gemma":[0.989755,0.007610855,0.0005102939,0.001102615,0.0007729087,0.0002482853],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001133925,0.0008741216,0.05198441,0.0002220592,0.0003027875,0.0004353451,0.001137278,0.7720107,0.009791165,0.002538014,0.002361201,0.157209],"study_design_scores_gemma":[0.00001463738,0.00007970315,0.0034551,0.000004540052,0.00001474397,0.000062209,0.0001263742,0.9928334,0.001316382,0.001765178,0.0003131601,0.00001460777],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6816853,0.0003095554,0.3145402,0.0004834259,0.0000270495,0.0001080025,0.0002823987,0.001952248,0.0006117884],"genre_scores_gemma":[0.8483446,0.0001621471,0.1494793,0.00005908938,0.00003708444,0.00007328029,0.000752707,0.00009658729,0.0009951915],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008605139,"threshold_uncertainty_score":0.02814478,"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."}}