{"id":"W2026695867","doi":"10.1145/1571941.1572089","title":"Integrating phrase inseparability in phrase-based model","year":2009,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Phrase; Computer science; Natural language processing; Artificial intelligence; Measure (data warehouse); Information retrieval; Data mining","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":[0.005742063,0.001211867,0.00169035,0.002528066,0.0008206223,0.002199554,0.002363135,0.001569613,0.002832097],"category_scores_gemma":[0.01507757,0.0006644767,0.001777186,0.002347815,0.001050459,0.007745174,0.001989229,0.003460953,0.002300107],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008157471,"about_ca_system_score_gemma":0.001297888,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002242041,"about_ca_topic_score_gemma":0.002618251,"domain_scores_codex":[0.9961825,0.001807051,0.0002434789,0.000604186,0.001014863,0.0001478884],"domain_scores_gemma":[0.9920239,0.005090613,0.0004298684,0.001355793,0.0009292577,0.0001704865],"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":[0.001090461,0.0005730182,0.005206706,0.001069784,0.001096802,0.0003110642,0.0007745618,0.2599531,0.04281374,0.1040774,0.009787417,0.5732459],"study_design_scores_gemma":[0.00004779443,0.0002882307,0.0008993901,0.00002835875,0.0002140507,0.0001885604,0.00004066595,0.9452936,0.004627534,0.04403607,0.004246381,0.00008923838],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01100147,0.0007495523,0.9849445,0.0004342182,0.00007022719,0.0001058613,0.0001631022,0.0007258055,0.001805282],"genre_scores_gemma":[0.4753251,0.001822861,0.5152539,0.0006947626,0.0006987801,0.0006392773,0.001400681,0.0004162218,0.003748394],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005742063,"threshold_uncertainty_score":0.03036726,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0304809453267647,"score_gpt":0.279191666179425,"score_spread":0.2487107208526603,"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."}}