{"id":"W2106854223","doi":"10.1111/j.1467-8640.2012.00436.x","title":"EXPLOITING SYNTACTIC, SEMANTIC, AND LEXICAL REGULARITIES IN LANGUAGE MODELING VIA DIRECTED MARKOV RANDOM FIELDS","year":2012,"lang":"en","type":"article","venue":"Computational Intelligence","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Language model; Perplexity; Computer science; Trigram; Probabilistic latent semantic analysis; Artificial intelligence; Natural language processing; Smoothing; Context (archaeology); Algorithm","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":[0.002017824,0.0007148996,0.0009417739,0.001401726,0.0005994879,0.0009404114,0.001570341,0.001282401,0.001311627],"category_scores_gemma":[0.006716916,0.0008339678,0.001535312,0.001302677,0.0008948654,0.002228394,0.0009718247,0.001544384,0.0005667325],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009866303,"about_ca_system_score_gemma":0.001533123,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008739644,"about_ca_topic_score_gemma":0.01324885,"domain_scores_codex":[0.9991971,0.0004334336,0.00003195733,0.0001676893,0.0001122851,0.00005753818],"domain_scores_gemma":[0.9952615,0.004051643,0.0002339062,0.000243871,0.0001369727,0.00007216477],"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":[0.00008818377,0.00006787772,0.001557249,0.00008661887,0.00006837977,0.0002479438,0.0001920392,0.8746275,0.00265795,0.06837356,0.001149758,0.05088306],"study_design_scores_gemma":[0.000005034767,0.000007589457,0.0000767235,0.000004418253,0.000006636038,0.00002213595,0.000004127406,0.9714516,0.0002353127,0.02789447,0.0002835738,0.000008286911],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01943275,0.0001816966,0.9789951,0.0001912251,0.00001912819,0.00002633738,0.0001574665,0.0004395684,0.0005567051],"genre_scores_gemma":[0.519005,0.0007374589,0.4752251,0.0002473671,0.00009389087,0.0003260337,0.0008265143,0.000325157,0.003213425],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008739644,"threshold_uncertainty_score":0.01737756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02393159905830056,"score_gpt":0.2965222874839078,"score_spread":0.2725906884256072,"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."}}