{"id":"W2904840750","doi":"10.48550/arxiv.1812.05272","title":"Towards a General-Purpose Linguistic Annotation Backend","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"National Science Foundation","keywords":"Computer science; Documentation; Annotation; Natural language processing; Upload; Transcription (linguistics); Process (computing); Artificial intelligence; Natural language; Linguistics; World Wide Web; Programming language","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":[0.004341427,0.002491117,0.00199934,0.002386047,0.001756898,0.009130679,0.005834224,0.002621976,0.02825907],"category_scores_gemma":[0.01313829,0.001715991,0.001922174,0.002246746,0.001282985,0.00850424,0.00712388,0.006448469,0.03906095],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00158793,"about_ca_system_score_gemma":0.002089957,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004609533,"about_ca_topic_score_gemma":0.005546187,"domain_scores_codex":[0.9970999,0.0004214688,0.0002584149,0.0008502557,0.001063181,0.0003067551],"domain_scores_gemma":[0.989984,0.002082879,0.0002611048,0.004225031,0.003014191,0.000432712],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.002121557,0.0006695766,0.002670332,0.001065664,0.0002311734,0.001341718,0.001829207,0.009345827,0.07366575,0.04270172,0.291232,0.5731255],"study_design_scores_gemma":[0.0002437137,0.0003186258,0.001921434,0.0004714331,0.0001913775,0.0008317361,0.0009049677,0.290626,0.1817913,0.09833812,0.4240797,0.0002814828],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004642474,0.0002338781,0.8155714,0.0007513434,0.0006356296,0.0003123053,0.004083077,0.1659922,0.007777657],"genre_scores_gemma":[0.07463098,0.0004716202,0.8246367,0.002404601,0.0005780191,0.0009393781,0.03933825,0.02446876,0.03253169],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02825907,"threshold_uncertainty_score":0.09453601,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05514361144347381,"score_gpt":0.2226009646340099,"score_spread":0.1674573531905361,"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."}}