{"id":"W2082225138","doi":"10.1145/1046456.1046458","title":"Extracting relational data from HTML repositories","year":2004,"lang":"en","type":"article","venue":"ACM SIGKDD Explorations Newsletter","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Set (abstract data type); Information retrieval; Data mining; Information extraction; Quality (philosophy)","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":[0.002320453,0.0009021745,0.001054459,0.01087782,0.0007095666,0.003061582,0.001582568,0.0009650076,0.001158987],"category_scores_gemma":[0.01986454,0.0007102127,0.001321234,0.009080175,0.0004567526,0.003221844,0.001894418,0.0009342164,0.001922132],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005577767,"about_ca_system_score_gemma":0.001329678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002656304,"about_ca_topic_score_gemma":0.004033797,"domain_scores_codex":[0.9969796,0.0004648877,0.000426762,0.0003990147,0.001579481,0.0001502828],"domain_scores_gemma":[0.9853367,0.00596384,0.001358155,0.004004272,0.003000018,0.000336957],"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":[0.0003661885,0.0003859211,0.04202877,0.001449787,0.0002699603,0.001771433,0.001875047,0.02556974,0.04613278,0.008904273,0.01632897,0.8549171],"study_design_scores_gemma":[0.0001323311,0.0004619152,0.03943773,0.0004592144,0.000741255,0.00374834,0.002586659,0.5392147,0.2862467,0.04468074,0.08199382,0.0002966518],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1676576,0.001911842,0.7773164,0.000998979,0.00007092505,0.0006377461,0.02084624,0.0251,0.005460272],"genre_scores_gemma":[0.2414712,0.001131223,0.7131192,0.0001314196,0.00007759978,0.0002576243,0.0415202,0.0008820247,0.001409501],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01087782,"threshold_uncertainty_score":0.01227188,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1297204511692792,"score_gpt":0.3005306323875561,"score_spread":0.170810181218277,"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."}}