{"id":"W2148362501","doi":"10.1017/s1351324904003560","title":"Correcting real-word spelling errors by restoring lexical cohesion","year":2005,"lang":"en","type":"article","venue":"Natural Language Engineering","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":180,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Spelling; Computer science; Cohesion (chemistry); Natural language processing; Lexicon; Artificial intelligence; Word (group theory); Context (archaeology); Precision and recall; Recall; Speech recognition; Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003714146,0.0002932264,0.0002484289,0.00019518,0.0001270568,0.0002002977,0.0008950278,0.000165557,0.00001134033],"category_scores_gemma":[0.0002943757,0.0002719601,0.00009110592,0.0005316759,0.00001652171,0.0009000982,0.0002581748,0.0008714293,0.0000265273],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00031639,"about_ca_system_score_gemma":0.00002626845,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001280037,"about_ca_topic_score_gemma":0.00001521602,"domain_scores_codex":[0.998158,0.00002959585,0.0003149663,0.0005049958,0.0004315398,0.0005609539],"domain_scores_gemma":[0.9990621,0.0001654766,0.000102411,0.0004776824,0.00005655045,0.0001357514],"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":[0.00001471384,0.00004012095,0.0001433605,0.00009228502,0.00002549757,0.0001668605,0.003186245,0.002293016,0.3500535,0.003082414,0.001718814,0.6391832],"study_design_scores_gemma":[0.0003432045,0.00004424497,0.0001001914,0.0003103466,0.00001326056,0.0002424661,0.0001235017,0.445722,0.5483044,0.00008124318,0.003772132,0.0009430088],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3763822,0.03882042,0.5752135,0.0007753409,0.00196365,0.0003344045,0.000003592611,0.005976592,0.0005303262],"genre_scores_gemma":[0.6607376,0.0000185994,0.3383821,0.0000779281,0.0003593745,0.000004804648,0.00000659769,0.00002848889,0.0003845784],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6382402,"threshold_uncertainty_score":0.9999732,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005958882391220053,"score_gpt":0.2512037126170917,"score_spread":0.2452448302258716,"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."}}