{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004410187,0.000120753,0.0001454959,0.0001037894,0.00004286057,0.00008063953,0.0005889636,0.00005387338,0.00001268505],"category_scores_gemma":[0.00009951314,0.0001043956,0.00004898458,0.0002940768,0.00001464709,0.0003851067,0.00005772843,0.000174529,0.00001373975],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000893398,"about_ca_system_score_gemma":0.000138332,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000100846,"about_ca_topic_score_gemma":0.0002004442,"domain_scores_codex":[0.9987848,0.00005154671,0.000291396,0.0004196607,0.0001926521,0.0002598989],"domain_scores_gemma":[0.999161,0.00003979319,0.00003998905,0.0006470552,0.00003680098,0.00007534609],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001537665,0.0004034152,0.003480223,0.00001403634,0.000002308609,0.00002558798,0.001190236,0.5519234,0.003833835,0.2527654,0.000212018,0.1861341],"study_design_scores_gemma":[0.0002851564,0.00002475716,0.0005539122,0.0000142263,5.976689e-7,8.394976e-7,0.00001113291,0.9687249,0.001105922,0.02914162,0.0000153089,0.0001216316],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2323107,0.00001279263,0.7598692,0.00153172,0.00003652745,0.00009605739,3.107665e-7,0.0001693438,0.005973374],"genre_scores_gemma":[0.7139004,3.366614e-7,0.2843781,0.001639564,0.00001209166,0.000004407197,5.57927e-7,0.00000221002,0.00006230261],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.4815896,"threshold_uncertainty_score":0.4257129,"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."}}