{"id":"W98931015","doi":"10.1038/npre.2008.2310.1","title":"Performance of the Charniak-Lease parser on biological text using different training corpora","year":2008,"lang":"en","type":"preprint","venue":"Nature Precedings","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Nature Conservancy of Canada; Carleton University; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Universities Space Research Association; Carleton University","keywords":"Computer science; Natural language processing; Parsing; Artificial intelligence; Text corpus","routes":{"ca_aff":true,"ca_fund":true,"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","research_integrity"],"consensus_categories":["research_integrity"],"category_scores_codex":[0.0003320116,0.000514556,0.0005877324,0.0001951818,0.0002237661,0.0001111383,0.003174103,0.001353461,0.000009870794],"category_scores_gemma":[0.0003322209,0.0003053213,0.000274862,0.0003623617,0.0002171346,0.0001844486,0.002525201,0.004063131,0.000002471467],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001674787,"about_ca_system_score_gemma":0.0001756976,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009117224,"about_ca_topic_score_gemma":6.12681e-7,"domain_scores_codex":[0.9973805,0.0001200862,0.0004570005,0.0009108885,0.0006920712,0.0004394337],"domain_scores_gemma":[0.9977062,0.0001441609,0.0008136484,0.001018174,0.000217622,0.0001002398],"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":[0.001034247,0.001755733,0.06637125,0.003453535,0.0007464371,0.000220496,0.03028936,0.002457003,0.2874431,0.06531383,0.007209695,0.5337053],"study_design_scores_gemma":[0.001334717,0.0009120863,0.02549263,0.01134711,0.0001225578,0.0004946478,0.00008306839,0.2779367,0.6421006,0.03431738,0.002193078,0.003665428],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9900175,0.00194211,0.004205693,0.0005067944,0.001506693,0.000600603,0.00001788587,0.000466517,0.0007362227],"genre_scores_gemma":[0.9353725,0.0003097139,0.06344691,0.0005280462,0.00022417,0.00002798442,0.000008651094,0.00002551507,0.00005652505],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5300399,"threshold_uncertainty_score":0.999943,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05159216876529966,"score_gpt":0.2861195937781131,"score_spread":0.2345274250128135,"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."}}