{"id":"W4402519936","doi":"10.1075/ijlcr.24010.paq","title":"The Core Metadata Schema for Learner Corpora (LC-meta)","year":2024,"lang":"en","type":"article","venue":"International Journal of Learner Corpus Research","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canarie","funders":"","keywords":"Metadata; Computer science; Schema (genetic algorithms); Information retrieval; Annotation; World Wide Web; Artificial intelligence","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.005630889,0.0001531158,0.0002381332,0.0006046751,0.0002672021,0.002856289,0.004669194,0.00009141042,0.00002984553],"category_scores_gemma":[0.001794235,0.00008753155,0.0003355251,0.0006216921,0.0003083482,0.001581188,0.0007715534,0.001300507,0.00003536641],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000179368,"about_ca_system_score_gemma":0.0005952488,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002121132,"about_ca_topic_score_gemma":0.000008780381,"domain_scores_codex":[0.9963048,0.0002028027,0.0005195072,0.0003328461,0.002244789,0.000395303],"domain_scores_gemma":[0.994261,0.001987436,0.0002195137,0.0004342144,0.002971855,0.0001259316],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001705451,0.00006708827,0.00007855381,0.0000424985,0.001509595,0.0006855854,0.0002806999,0.00002751922,0.008183637,0.6334301,0.06976648,0.2857577],"study_design_scores_gemma":[0.0003753293,0.0003658617,0.00002092329,0.0002004556,0.00007908726,0.00116084,0.00008715708,0.01714152,0.01622284,0.2923244,0.6718138,0.0002078433],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002073956,0.07600455,0.880843,0.03692869,0.002928662,0.000374628,0.00002206485,0.0002140903,0.0006103545],"genre_scores_gemma":[0.6005394,0.002047013,0.3647136,0.0004765824,0.002158175,0.000087344,0.00001375313,0.0000793716,0.02988472],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6020473,"threshold_uncertainty_score":0.9981788,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2257686942419148,"score_gpt":0.4712745965182499,"score_spread":0.245505902276335,"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."}}