{"id":"W4310183442","doi":"10.18280/isi.270507","title":"A Hybrid Approach for Web Pages Classification","year":2022,"lang":"en","type":"article","venue":"Ingénierie des systèmes d information","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Web page; Cover (algebra); The Internet; Information retrieval; World Wide Web; Artificial intelligence; Data mining; Engineering","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007056615,0.000104061,0.0001337781,0.0002726982,0.0006609806,0.0003770874,0.0006706048,0.00002128947,0.00001134596],"category_scores_gemma":[0.0001090523,0.0001059935,0.00008252462,0.0004436248,0.00003451825,0.002719297,0.0002191624,0.00009234867,0.00001735119],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001791744,"about_ca_system_score_gemma":0.0001012089,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001033183,"about_ca_topic_score_gemma":3.419251e-7,"domain_scores_codex":[0.9989078,0.00006742153,0.0003637591,0.0001691384,0.0002824349,0.000209436],"domain_scores_gemma":[0.9991025,0.00005281253,0.0002562003,0.0004263073,0.0001154802,0.00004664869],"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":[0.00004095074,0.0001554426,0.0008813833,0.0003918899,0.0001341893,0.000001728818,0.009192687,0.009613593,0.001418415,0.1163474,0.04840716,0.8134152],"study_design_scores_gemma":[0.0002758178,0.00007932899,0.0008782884,0.000006255534,0.00001434719,0.00003064614,0.001171045,0.9519027,0.000234471,0.001925832,0.04330341,0.0001778787],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01417695,0.00004600024,0.9781007,0.0001626617,0.0001592936,0.0002448834,0.0001005316,0.0002492195,0.006759748],"genre_scores_gemma":[0.9383064,0.000006279735,0.06013837,0.0002094785,0.00003616199,0.000434239,0.0007703709,0.000005295378,0.00009338869],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9422891,"threshold_uncertainty_score":0.5083795,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02966973419881574,"score_gpt":0.2344146807610414,"score_spread":0.2047449465622257,"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."}}