{"id":"W162289556","doi":"","title":"Mining Multiple Web Sources Using Non-Deterministic Finite State Automata","year":2012,"lang":"en","type":"article","venue":"Scholarship at UWindsor (University of Windsor)","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Computer science; Information extraction; Web page; Metadata; Information retrieval; Deterministic finite automaton; Automaton; Data extraction; Finite-state machine; Object (grammar); Data mining; Theoretical computer science; Algorithm; World Wide Web; Artificial intelligence","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":[0.001326269,0.001080054,0.0009291632,0.003310215,0.001207175,0.002067713,0.001665068,0.001009992,0.001322397],"category_scores_gemma":[0.007762762,0.0008818221,0.002371775,0.001871626,0.0007741119,0.002282503,0.001486717,0.0007740931,0.0005069765],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001238407,"about_ca_system_score_gemma":0.001851537,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01125771,"about_ca_topic_score_gemma":0.01839634,"domain_scores_codex":[0.9981666,0.0003453733,0.0002557246,0.0006245948,0.0005057156,0.0001019993],"domain_scores_gemma":[0.9916728,0.006343238,0.0006259956,0.000599306,0.0006614433,0.0000972939],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004042912,0.0004301671,0.04103917,0.0006872904,0.0005641476,0.002509338,0.001115241,0.6757268,0.0164374,0.01736861,0.001838439,0.241879],"study_design_scores_gemma":[0.00001825482,0.00003303589,0.001200466,0.00002886092,0.00007112557,0.0001414192,0.0001084999,0.9798337,0.006316156,0.011187,0.001040503,0.00002099849],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09221207,0.0001755043,0.9001873,0.0002490462,0.00002420188,0.0003379061,0.001189858,0.004142338,0.001481902],"genre_scores_gemma":[0.5219235,0.0002111802,0.4727271,0.00009360879,0.00001793482,0.0005664099,0.002637415,0.0001759312,0.001646995],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01125771,"threshold_uncertainty_score":0.02238435,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03266818413818435,"score_gpt":0.2337784940834311,"score_spread":0.2011103099452468,"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."}}