{"id":"W2147979708","doi":"10.1145/967900.968022","title":"Framework for mining web content outliers","year":2004,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Computer science; Outlier; Content (measure theory); Web mining; Information retrieval; World Wide Web; Web service; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001658064,0.0000725078,0.000111568,0.00006185444,0.00009328013,0.0001376504,0.0005252208,0.00004058516,0.00001076499],"category_scores_gemma":[0.0001352128,0.00005865692,0.00008319135,0.000204261,0.00001841588,0.0002018379,0.00009072196,0.00004800884,0.00004333397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000226484,"about_ca_system_score_gemma":0.00005537229,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002309022,"about_ca_topic_score_gemma":0.0000106464,"domain_scores_codex":[0.9992977,0.000005613077,0.0001300188,0.0002531646,0.0001146888,0.0001988274],"domain_scores_gemma":[0.999364,0.00009545577,0.00003937515,0.0003886351,0.00003972822,0.00007281053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000002912669,0.0000471047,0.0004772138,0.000006144797,0.00005718298,0.000005302209,0.000824628,0.0003480254,0.0002903118,0.9577336,0.002396365,0.03781125],"study_design_scores_gemma":[0.007884375,0.001389283,0.002407884,0.0006606527,0.0002694213,0.00005588859,0.01002348,0.3554679,0.02427313,0.4352931,0.1591993,0.003075593],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01135573,0.00003162233,0.9827929,0.004164945,0.0001841189,0.00004848501,0.000004060793,0.0001500324,0.001268112],"genre_scores_gemma":[0.3113546,0.000003309562,0.6873003,0.0008719257,0.0000467478,0.000009250703,0.000002881978,0.000003651848,0.0004073685],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5224404,"threshold_uncertainty_score":0.239196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0725405980701763,"score_gpt":0.2891242805934112,"score_spread":0.2165836825232349,"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."}}