{"id":"W160940076","doi":"10.63317/3vk8aoyuzdat","title":"Targeting Chinese Nominal Compounds in Corpora","year":2008,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Natural language processing; Artificial intelligence; Pipeline (software); Lexicon; Sentence; Morpheme; Set (abstract data type); Speech recognition","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":[0.003161211,0.001593855,0.001410163,0.005401108,0.00253269,0.002509836,0.001512003,0.001466828,0.008545115],"category_scores_gemma":[0.008725627,0.000887328,0.0008115971,0.006421947,0.0009953992,0.003506901,0.002448947,0.001070777,0.007032413],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001710579,"about_ca_system_score_gemma":0.003278609,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01185155,"about_ca_topic_score_gemma":0.01748343,"domain_scores_codex":[0.9969845,0.0005816898,0.0002797309,0.001168643,0.0006815309,0.0003039261],"domain_scores_gemma":[0.9954361,0.00146035,0.0003652359,0.000998996,0.00158411,0.0001551493],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009109543,0.0003446453,0.02321798,0.002955363,0.0002032638,0.002177632,0.002710617,0.007726329,0.1904754,0.02335276,0.1040062,0.6419189],"study_design_scores_gemma":[0.0003426216,0.0004954558,0.0488,0.0003452182,0.0005000978,0.00266563,0.002699875,0.1025208,0.3436321,0.01766134,0.4800202,0.0003165434],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.340136,0.006285681,0.5037736,0.003278059,0.0007725915,0.002455158,0.0554485,0.03083241,0.05701808],"genre_scores_gemma":[0.3151908,0.001888533,0.5507244,0.0005583308,0.0002527482,0.001404512,0.1122757,0.002185893,0.01551906],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01185155,"threshold_uncertainty_score":0.02858621,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01310557270870183,"score_gpt":0.2589636730924225,"score_spread":0.2458581003837206,"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."}}