{"id":"W2027081355","doi":"10.3115/1699648.1699697","title":"Character-level analysis of semi-structured documents for set expansion","year":2009,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Association of Canadian Universities for Research in Astronomy","keywords":"Set (abstract data type); Seal (emblem); Computer science; Character (mathematics); Independence (probability theory); Artificial intelligence; Character encoding; Natural language processing; Programming language; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001487822,0.0001048527,0.000228732,0.0002960608,0.0000420687,0.00006559262,0.0006610136,0.00006755533,0.00001848489],"category_scores_gemma":[0.00003672206,0.00007880011,0.0001198792,0.0008241507,0.00001109858,0.000436713,0.00007245778,0.00005546563,7.592533e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002297258,"about_ca_system_score_gemma":0.00002098237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001792011,"about_ca_topic_score_gemma":0.000006724903,"domain_scores_codex":[0.9991312,0.00001591591,0.0002165956,0.000270193,0.0002030497,0.0001630297],"domain_scores_gemma":[0.9992375,0.00003583348,0.0001329384,0.0004221409,0.000130004,0.00004156222],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00004462904,0.0001054467,0.0008539159,0.00004768994,0.0003759447,0.000006768965,0.001539683,0.00002478854,0.4205226,0.06811631,0.003919996,0.5044423],"study_design_scores_gemma":[0.0004238933,0.0002203028,0.01448657,0.00003822573,0.0002244641,0.000004811292,0.00001653958,0.05492796,0.8504292,0.07827121,0.000573837,0.000382917],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03551205,0.0001368705,0.9632526,0.0004712611,0.00005773983,0.0001612094,0.00001875976,0.0002746909,0.000114834],"genre_scores_gemma":[0.5747963,0.000002449172,0.424534,0.0004026822,0.00001042669,0.000003513956,0.00002035183,0.000002008904,0.0002283083],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5392842,"threshold_uncertainty_score":0.3213375,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02246672517775284,"score_gpt":0.3147622950714681,"score_spread":0.2922955698937153,"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."}}