{"id":"W17684","doi":"10.1111/j.2042-7158.1977.tb11305.x","title":"Word pairs in language modeling for information retrieval","year":2004,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Bigram; Vector space model; Computer science; Word (group theory); Language model; Natural language processing; Artificial intelligence; Constraint (computer-aided design); Term Discrimination; Adjacency list; Information retrieval; Visual Word; Algorithm; Mathematics","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":[0.003355091,0.001339067,0.001889061,0.002301365,0.001042746,0.002549868,0.001722207,0.001711774,0.007045361],"category_scores_gemma":[0.01254387,0.0006470784,0.001732095,0.003422169,0.00078125,0.007051956,0.001715669,0.002207717,0.003658466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001271229,"about_ca_system_score_gemma":0.00134301,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005265391,"about_ca_topic_score_gemma":0.004304782,"domain_scores_codex":[0.9967598,0.002163726,0.0001692355,0.0004161041,0.0003769082,0.0001142247],"domain_scores_gemma":[0.9962185,0.002910208,0.0001794072,0.0003070984,0.0003128607,0.00007192469],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003056397,0.0002276298,0.001707366,0.0008705329,0.000438417,0.0003300948,0.0009555678,0.2307088,0.003548905,0.4037711,0.01895343,0.3381824],"study_design_scores_gemma":[0.00003246954,0.00005637345,0.0002467552,0.00005266543,0.00008202162,0.0001146136,0.000124775,0.6654794,0.0009818305,0.321215,0.01156277,0.00005141762],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007965537,0.00459344,0.9807408,0.001322092,0.0003222396,0.0001352062,0.0004076865,0.0009563682,0.003556611],"genre_scores_gemma":[0.3541711,0.007004868,0.6198763,0.001012522,0.001120886,0.001342145,0.002335813,0.0007862967,0.01235009],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007045361,"threshold_uncertainty_score":0.02356905,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02039087438294956,"score_gpt":0.2508621982835105,"score_spread":0.2304713239005609,"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."}}