{"id":"W2074126326","doi":"10.1145/2213836.2213912","title":"Automatic web-scale information extraction","year":2012,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Scale (ratio); Extraction (chemistry); Information extraction; Information retrieval; World Wide Web; Chromatography; Geography","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":[0.001842135,0.001052234,0.001068383,0.002968021,0.0009319559,0.002859146,0.001056668,0.001061927,0.007828068],"category_scores_gemma":[0.007879726,0.000643889,0.001111641,0.004147124,0.0005179131,0.003588748,0.003222962,0.001245713,0.009169497],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004542423,"about_ca_system_score_gemma":0.000895241,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001529549,"about_ca_topic_score_gemma":0.002588066,"domain_scores_codex":[0.99823,0.0002992116,0.0001689534,0.0003671594,0.000842334,0.00009221179],"domain_scores_gemma":[0.9950412,0.00204765,0.0002197052,0.00158631,0.0009908361,0.0001144211],"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":[0.000501744,0.0003574795,0.004810782,0.001170312,0.000228929,0.001257151,0.0008636867,0.006285354,0.09071698,0.02001044,0.1189241,0.754873],"study_design_scores_gemma":[0.0002006424,0.0001453915,0.009068615,0.000399316,0.0002190894,0.002279092,0.000849931,0.278426,0.252499,0.07677985,0.3789129,0.0002201732],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01287939,0.0006461523,0.9019634,0.0007780301,0.0001351876,0.0004090622,0.006126782,0.06930766,0.007754353],"genre_scores_gemma":[0.07970366,0.0005874535,0.8951151,0.0004792241,0.00006708245,0.0003558366,0.01422953,0.0035807,0.005881347],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007828068,"threshold_uncertainty_score":0.02618748,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01052818257282117,"score_gpt":0.2457080567536289,"score_spread":0.2351798741808077,"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."}}