{"id":"W2155085300","doi":"10.1109/hicss.2008.53","title":"An Examination of Genre Attributes for Web Page Classification","year":2008,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Web page; Set (abstract data type); Information retrieval; Identification (biology); World Wide Web; Recall; Precision and recall","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":[0.0005013195,0.0000565495,0.00008024336,0.00006574737,0.0001217839,0.00001728182,0.000295289,0.00005717626,0.00002107166],"category_scores_gemma":[0.00006297715,0.00005129018,0.00003484879,0.0002261315,0.00002685344,0.0004111712,0.00002187944,0.00003575945,0.00001068265],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001967465,"about_ca_system_score_gemma":0.00004948181,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001819051,"about_ca_topic_score_gemma":0.00000150074,"domain_scores_codex":[0.9993198,0.00007005782,0.0001783378,0.0001847306,0.0001325476,0.0001145057],"domain_scores_gemma":[0.9993023,0.00009592275,0.00008769635,0.0002707737,0.0001959528,0.00004734474],"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":[0.00000887428,0.0001605089,0.00637047,0.00003595642,0.000009572142,0.000001160411,0.001324053,0.00002773279,0.2686871,0.6588916,0.001103525,0.06337943],"study_design_scores_gemma":[0.0003828858,0.0001834002,0.2689103,0.000006025618,0.000004055675,0.000008485316,0.0001202982,0.6106579,0.1168075,0.001012725,0.001745716,0.0001607419],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1983922,0.00002208527,0.8000523,0.0003309383,0.00008248141,0.000148389,0.000007904335,0.0001086135,0.0008551795],"genre_scores_gemma":[0.9466976,0.000007920488,0.05270319,0.00006390784,0.00002973235,0.00001674236,0.0000418643,0.000003156733,0.0004358841],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7483054,"threshold_uncertainty_score":0.2091553,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1191895850919988,"score_gpt":0.3147126104816039,"score_spread":0.1955230253896051,"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."}}