{"id":"W59948950","doi":"10.5220/0001837706460653","title":"CLASSIFYING WEB PAGES BY GENRE - A Distance Function Approach","year":2009,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Web page; Similarity (geometry); Information retrieval; Centroid; Byte; Set (abstract data type); Preprocessor; Function (biology); World Wide Web; Artificial intelligence; Image (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":[0.0002737039,0.00009845964,0.0000934059,0.00003540143,0.0001678627,0.0001492069,0.000322844,0.00006605692,0.00002217498],"category_scores_gemma":[0.00001619068,0.00008262472,0.00004222498,0.0003299751,0.00001443248,0.0004117897,0.00003865506,0.0001390535,0.0000538264],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000319321,"about_ca_system_score_gemma":0.00002572972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001229276,"about_ca_topic_score_gemma":7.614694e-7,"domain_scores_codex":[0.9990825,0.000055171,0.0001418243,0.0003068072,0.0001862358,0.0002274751],"domain_scores_gemma":[0.9995402,0.00002276018,0.00004869716,0.0002744432,0.00003288902,0.00008105765],"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":[0.00001873422,0.000140871,0.001275833,0.00001755488,0.00001305193,0.000003462966,0.0003656657,0.00005870177,0.01197266,0.7525041,0.0535772,0.1800522],"study_design_scores_gemma":[0.0008842386,0.0003001855,0.009574832,0.00002877759,0.0000161303,0.00002572853,0.0003462104,0.7275227,0.007923227,0.01735692,0.2351107,0.000910326],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001134187,0.000409462,0.9641888,0.001975491,0.0001594087,0.0000797262,0.000001648602,0.0003939371,0.03165741],"genre_scores_gemma":[0.9648324,0.00002015589,0.03001129,0.001337912,0.00004601968,0.000004911241,0.00001260896,0.000003160114,0.003731599],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9636981,"threshold_uncertainty_score":0.3369338,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03294272366160689,"score_gpt":0.2517075866206458,"score_spread":0.2187648629590389,"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."}}