{"id":"W2098181468","doi":"10.1109/hicss.2005.114","title":"Automatic Identification of Home Pages on the Web","year":2005,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Identification (biology); Computer science; World Wide Web; Web page; Biology","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":[0.0004242469,0.0002948503,0.0004404297,0.005512543,0.0004527662,0.001357408,0.0003472884,0.0005282444,0.00202577],"category_scores_gemma":[0.002117557,0.0001882652,0.0002953387,0.001849285,0.0002250483,0.001206912,0.0005619791,0.000436025,0.001728307],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002482142,"about_ca_system_score_gemma":0.000292673,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002965024,"about_ca_topic_score_gemma":0.005142911,"domain_scores_codex":[0.9996141,0.00008116393,0.00002512544,0.00008516265,0.0001440765,0.0000503841],"domain_scores_gemma":[0.9984319,0.0004921959,0.0001927503,0.0002079458,0.0005756733,0.0000995697],"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":[0.0007962265,0.0004196741,0.1736936,0.0005891396,0.0001289054,0.001833204,0.001135307,0.003187215,0.04811512,0.003842104,0.02811001,0.7381495],"study_design_scores_gemma":[0.00006566491,0.0002505633,0.5271004,0.0002973563,0.0002396469,0.004556507,0.002757072,0.3221065,0.08394375,0.01169125,0.0468754,0.0001158631],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.8994268,0.002168907,0.06190924,0.0002346696,0.0001471911,0.0003533649,0.005381771,0.0061322,0.02424587],"genre_scores_gemma":[0.906306,0.0007988726,0.07426788,0.00007676402,0.00009063361,0.000087602,0.008886081,0.0002841151,0.009202112],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.005512543,"threshold_uncertainty_score":0.006776929,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0194470054804127,"score_gpt":0.2397336240570869,"score_spread":0.2202866185766742,"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."}}