{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003481423,0.001039346,0.00107481,0.01029752,0.0008877596,0.003534628,0.00117655,0.001520371,0.002667107],"category_scores_gemma":[0.01519294,0.0002321348,0.001269921,0.006072009,0.0006311472,0.003526688,0.001262789,0.001624759,0.003293227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001369217,"about_ca_system_score_gemma":0.0007975833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002761089,"about_ca_topic_score_gemma":0.002351373,"domain_scores_codex":[0.9974697,0.0007518492,0.0003080608,0.0004355637,0.0008881819,0.0001466251],"domain_scores_gemma":[0.9905499,0.004565774,0.0007592562,0.001174074,0.002432333,0.0005186077],"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":[0.0008723377,0.001302239,0.1106241,0.0003683878,0.0003337947,0.0001969842,0.0007710168,0.02416444,0.01487689,0.005441618,0.006702925,0.8343453],"study_design_scores_gemma":[0.00005424732,0.0004755455,0.0464721,0.00007321143,0.00009925837,0.0005829061,0.0008242282,0.9252712,0.009713065,0.01075826,0.005577507,0.00009849768],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5446909,0.00142149,0.4370699,0.0008668406,0.0003256318,0.0009431049,0.003012147,0.002592305,0.009077623],"genre_scores_gemma":[0.691326,0.0004924528,0.298172,0.0001182706,0.0001477447,0.0003785017,0.003711555,0.0001870912,0.005466338],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01029752,"threshold_uncertainty_score":0.0184117,"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."}}