{"id":"W1030963941","doi":"10.1163/9789401206884_013","title":"Corpus linguistics and language documentation: challenges for collaboration","year":2011,"lang":"en","type":"book-chapter","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Documentation; Corpus linguistics; Linguistics; Computer science; Text corpus; Applied linguistics; Natural language processing; Philosophy; Programming language","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.1017109,0.0008880587,0.002321022,0.01106679,0.009708311,0.0397744,0.006363507,0.008134305,0.007157788],"category_scores_gemma":[0.1512905,0.001472371,0.0009119318,0.01664338,0.02762683,0.06064387,0.0197253,0.01341201,0.00311806],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.008505989,"about_ca_system_score_gemma":0.02717377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006171937,"about_ca_topic_score_gemma":0.006448732,"domain_scores_codex":[0.8942908,0.08178619,0.004857571,0.00459878,0.01322842,0.001238207],"domain_scores_gemma":[0.7287289,0.2244194,0.004178811,0.02588386,0.01268333,0.00410577],"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":[0.00002000702,0.00004283614,0.0004811558,0.0006594538,0.00002141402,0.0001751638,0.01992378,0.000560133,0.000179336,0.7188155,0.04055689,0.2185644],"study_design_scores_gemma":[0.00002020133,0.00001595212,0.0002781816,0.002703755,0.000009047949,0.0002562043,0.01121015,0.00113705,0.0001828874,0.6516122,0.3325238,0.00005063776],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"other","genre_scores_codex":[0.008768062,0.1644137,0.3097591,0.3795748,0.004227574,0.0003697799,0.0003249715,0.001384283,0.1311778],"genre_scores_gemma":[0.1852569,0.1560949,0.5713736,0.02805096,0.00843735,0.002291377,0.001157793,0.002993837,0.04434323],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.1017109,"threshold_uncertainty_score":0.5379049,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02729081510545165,"score_gpt":0.2927130182598712,"score_spread":0.2654222031544195,"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."}}