{"id":"W4292510811","doi":"10.6087/kcse.285","title":"Improving Journal Article Tag Suite for multilingual articles","year":2022,"lang":"en","type":"article","venue":"Science Editing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; Canadian Medical Association","keywords":"Suite; Computer science; Variety (cybernetics); Set (abstract data type); World Wide Web; Information retrieval; Library science; Data science; Political science; Artificial intelligence; Programming language; Law","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.003808335,0.00009552127,0.00009522661,0.0002324995,0.002505915,0.0009173088,0.002070938,0.00001670246,0.00001425935],"category_scores_gemma":[0.001906878,0.00008846947,0.00005402832,0.001105781,0.0001783642,0.001830652,0.001126978,0.0003535216,0.000002177135],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001865887,"about_ca_system_score_gemma":0.0003572901,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000144754,"about_ca_topic_score_gemma":0.000001987374,"domain_scores_codex":[0.9979098,0.00004148987,0.0002530789,0.0004109345,0.0007723309,0.0006123603],"domain_scores_gemma":[0.9989348,0.0001611159,0.0002247068,0.0002787406,0.000264118,0.0001365009],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000005059644,0.00003885585,0.0003757046,0.00001199874,0.000001798112,0.00002853475,0.00298236,0.0001697294,0.5211667,0.01107191,0.0003744203,0.4637729],"study_design_scores_gemma":[0.0003126356,0.0001686497,0.00006188315,0.00001683029,0.000004107592,0.0002730238,0.001552356,0.3796363,0.5812221,0.03581777,0.0006918212,0.0002425795],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3466982,0.0003511261,0.6490881,0.001070099,0.002062115,0.000196241,0.000003171603,0.0004677524,0.00006318632],"genre_scores_gemma":[0.6197065,1.87449e-7,0.3796176,0.0002280107,0.0003911945,0.00002013355,1.868062e-7,0.000005297979,0.00003089798],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.4635303,"threshold_uncertainty_score":0.9987927,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01536239155432406,"score_gpt":0.2954643337713264,"score_spread":0.2801019422170024,"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."}}