{"id":"W2077255111","doi":"10.3115/1118627.1118636","title":"Acquiring collocations for lexical choice between near-synonyms","year":2002,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto; University of Pennsylvania","keywords":"Collocation (remote sensing); Computer science; Synonym (taxonomy); Natural language processing; Artificial intelligence; Lexical density; Word (group theory); Linguistics; Task (project management); Lexical item","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002069265,0.0007155539,0.0008929686,0.003466479,0.001106099,0.001685975,0.001493534,0.00120584,0.008213305],"category_scores_gemma":[0.01869494,0.0006535536,0.0006735367,0.002461699,0.001123847,0.006995547,0.002796134,0.00156021,0.003297496],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006741909,"about_ca_system_score_gemma":0.001052675,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002754492,"about_ca_topic_score_gemma":0.005725595,"domain_scores_codex":[0.997179,0.0006899465,0.000318225,0.00115309,0.0005121075,0.0001476194],"domain_scores_gemma":[0.9846631,0.00983435,0.0009415072,0.002171378,0.001966753,0.0004229038],"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":[0.0007134103,0.0007113612,0.08304092,0.0009344034,0.0002424831,0.002108727,0.007861663,0.01078917,0.09744798,0.02216377,0.008323559,0.7656627],"study_design_scores_gemma":[0.000335045,0.0008201972,0.1694164,0.0005491984,0.0004044673,0.005321119,0.01448179,0.4105588,0.1190025,0.2075365,0.07103167,0.0005422843],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5360507,0.000361845,0.4447891,0.0005514931,0.00008392101,0.0003666058,0.003882597,0.002827926,0.01108583],"genre_scores_gemma":[0.7819096,0.0001246724,0.2110966,0.0001012065,0.0000241083,0.0002061716,0.004864415,0.0002690936,0.001404177],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008213305,"threshold_uncertainty_score":0.02747625,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04991608310263382,"score_gpt":0.3119821999308962,"score_spread":0.2620661168282623,"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."}}