{"id":"W2064514861","doi":"10.1142/s021848850400317x","title":"TEXTUAL-BASED CLUSTERING OF WEB DOCUMENTS","year":2004,"lang":"en","type":"article","venue":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Cluster analysis; Computer science; Information retrieval; Schematic; Fuzzy clustering; Directory; Web page; HTML element; Document clustering; Data mining; Web mining; Feature (linguistics); World Wide Web; 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":[0.001786507,0.0007565885,0.0008142604,0.009386598,0.001006531,0.002330581,0.001252601,0.001062709,0.00240019],"category_scores_gemma":[0.01088152,0.0002860634,0.0008883723,0.006979939,0.0005883536,0.001944919,0.0006670926,0.0005460251,0.002203993],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001333687,"about_ca_system_score_gemma":0.001020385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003895962,"about_ca_topic_score_gemma":0.005035459,"domain_scores_codex":[0.9970638,0.0005823976,0.0003705731,0.0006428839,0.001216348,0.0001239092],"domain_scores_gemma":[0.994493,0.001583581,0.000476205,0.0005515589,0.002760671,0.0001350431],"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":[0.000890999,0.0005289171,0.01657113,0.002423671,0.0003116672,0.0004908837,0.001451672,0.04551757,0.08154033,0.01317344,0.008719961,0.8283798],"study_design_scores_gemma":[0.0001415023,0.0005483649,0.04135919,0.000464753,0.0003745795,0.00179096,0.00204236,0.735751,0.139139,0.02832002,0.04977795,0.000290357],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1133474,0.00158299,0.8712491,0.000381996,0.0001742047,0.0008127193,0.003130324,0.003148431,0.006172768],"genre_scores_gemma":[0.2893834,0.0005938305,0.6981568,0.0001435574,0.0001621313,0.0004349544,0.005779389,0.000326807,0.005019155],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009386598,"threshold_uncertainty_score":0.009676576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01590388508604498,"score_gpt":0.2784247356577021,"score_spread":0.2625208505716571,"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."}}