{"id":"W4230874668","doi":"10.1007/978-1-4614-6170-8_100858","title":"Web Crawling","year":2014,"lang":"en","type":"book-chapter","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary","funders":"","keywords":"Crawling; Computer science; World Wide Web; Medicine; Anatomy","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":[0.0009024483,0.001592852,0.001109437,0.007775414,0.001399386,0.003422517,0.001580312,0.001193573,0.01544819],"category_scores_gemma":[0.003619927,0.0008820028,0.0009912419,0.007977644,0.0006145278,0.004438681,0.00215885,0.001536168,0.02559851],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008563265,"about_ca_system_score_gemma":0.001400051,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002358608,"about_ca_topic_score_gemma":0.005160396,"domain_scores_codex":[0.9985235,0.0001369205,0.0001008159,0.0003062039,0.0008419041,0.00009063315],"domain_scores_gemma":[0.9980129,0.0004263864,0.0001045689,0.0007348438,0.0006354497,0.00008583313],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001821489,0.00007862091,0.0004611563,0.0004131922,0.0000319766,0.00007978525,0.000117846,0.0009527672,0.003172172,0.01105752,0.1604932,0.8231235],"study_design_scores_gemma":[0.00001413217,0.00005049933,0.002035157,0.0005094637,0.00008442548,0.001754203,0.0001332458,0.02181667,0.01790778,0.04360752,0.9120272,0.00005967875],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01686501,0.02888878,0.5212587,0.004517545,0.002569847,0.0008398642,0.007295329,0.04905415,0.3687108],"genre_scores_gemma":[0.05447173,0.02633998,0.4163847,0.00206263,0.001405282,0.0004659097,0.02333433,0.009373492,0.4661621],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01544819,"threshold_uncertainty_score":0.05167931,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01957269536115172,"score_gpt":0.2193976681005148,"score_spread":0.1998249727393631,"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."}}