{"id":"W2781697036","doi":"10.1007/978-3-319-73706-5_13","title":"In-Memory Distributed Training of Linear-Chain Conditional Random Fields with an Application to Fine-Grained Named Entity Recognition","year":2018,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Banting and Best Diabetes Centre, University of Toronto; Bundesministerium für Bildung und Forschung","keywords":"Computer science; Conditional random field; Scalability; Named-entity recognition; Task (project management); Sequence labeling; Feature (linguistics); Sequence (biology); Artificial intelligence; Pattern recognition (psychology); Database","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009899373,0.0003067776,0.0004504798,0.0005861289,0.0001242009,0.0001081912,0.00134195,0.000242713,0.00003147067],"category_scores_gemma":[0.0001393582,0.0002859742,0.00006109817,0.0005777943,0.0003349222,0.0005376972,0.0003099481,0.0004390174,0.000009927956],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001523218,"about_ca_system_score_gemma":0.0003872527,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005972806,"about_ca_topic_score_gemma":0.0006779576,"domain_scores_codex":[0.9972357,0.00005067277,0.0005371499,0.001116503,0.0006831601,0.0003768289],"domain_scores_gemma":[0.9980473,0.0003378606,0.0002798837,0.0008241346,0.000367846,0.0001429664],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002215854,0.0001397191,0.0001015822,0.0001146663,0.00002679281,0.00006468479,0.008844444,0.2611573,0.0007964732,0.005847997,0.00002721808,0.7226575],"study_design_scores_gemma":[0.0009975801,0.0003433922,0.000111816,0.0002899029,0.000006202071,0.00002891576,0.000001366373,0.8698863,0.001539229,0.1263179,0.00008046849,0.0003969262],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007276103,0.00001514009,0.9908141,0.0006527576,0.000355325,0.0006109588,0.00003681332,0.00006741843,0.0001714136],"genre_scores_gemma":[0.6051444,0.000001493271,0.3936192,0.0006978223,0.0003619312,0.00003243788,0.00009548065,0.00001417286,0.00003301203],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7222606,"threshold_uncertainty_score":0.9999592,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02398787561735934,"score_gpt":0.2561097958476444,"score_spread":0.2321219202302851,"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."}}