{"id":"W2110780226","doi":"10.1145/1031453.1031458","title":"Probabilistic models for focused web crawling","year":2004,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Mitacs","keywords":"CRFS; Crawling; Computer science; Conditional random field; Hidden Markov model; Web crawler; Probabilistic logic; Context (archaeology); Relevance (law); Artificial intelligence; Web page; Machine learning; Information retrieval; Data mining; World Wide Web","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0001809545,0.00007208861,0.0001040583,0.00006251258,0.00009119912,0.0001418294,0.000526754,0.0000258511,0.000004320128],"category_scores_gemma":[0.00005136069,0.00005804471,0.00006924535,0.0002253588,0.00001433154,0.0004123337,0.00008893119,0.00003413849,0.0000241066],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002531919,"about_ca_system_score_gemma":0.00009767306,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003497794,"about_ca_topic_score_gemma":0.00003152163,"domain_scores_codex":[0.9992788,0.000007434523,0.0001316334,0.0002836322,0.000112695,0.000185801],"domain_scores_gemma":[0.9993896,0.00005512786,0.00002933769,0.0004154501,0.00005469858,0.00005580628],"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":[0.000003667182,0.0001105114,0.00001572174,0.00002848168,0.00003985649,0.000004128738,0.0004460265,0.1215669,0.00126328,0.851074,0.0007363201,0.0247111],"study_design_scores_gemma":[0.000363364,0.00003481131,0.000003702689,0.00001102262,0.000009669058,0.000001459414,0.000009662352,0.8530109,0.0009054922,0.1450061,0.0005428778,0.0001009731],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005660754,0.00001780369,0.9906815,0.001248732,0.00005709464,0.00008891656,0.000004258619,0.0001966984,0.002044259],"genre_scores_gemma":[0.6617559,0.000001300644,0.337736,0.0001547639,0.00002964726,0.00001766122,0.000003392564,0.000003853642,0.0002975781],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7314439,"threshold_uncertainty_score":0.2366995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03819472388232283,"score_gpt":0.2513862193380534,"score_spread":0.2131914954557306,"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."}}