{"id":"W2160825952","doi":"10.1145/1008992.1009024","title":"Dependence language model for information retrieval","year":2004,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":261,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Language model; Term (time); Smoothing; Linkage (software); Divergence-from-randomness model; Artificial intelligence; Probabilistic logic; Independence (probability theory); Graph; Sentence; Term Discrimination; Axiom; Natural language processing; Theoretical computer science; Visual Word; Mathematics; Image retrieval","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":[0.002306017,0.0008487065,0.001182874,0.002180255,0.0006049019,0.0015316,0.002291122,0.001322519,0.004899609],"category_scores_gemma":[0.008259196,0.0005337208,0.001565557,0.002124184,0.000837755,0.005835143,0.001087348,0.001996858,0.003029622],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001795441,"about_ca_system_score_gemma":0.001141232,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006065931,"about_ca_topic_score_gemma":0.003293014,"domain_scores_codex":[0.9980962,0.0007543518,0.0001251166,0.0003672727,0.0005280552,0.0001290545],"domain_scores_gemma":[0.9965218,0.002226819,0.0002242516,0.0004670895,0.0004818903,0.00007820795],"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":[0.0002374779,0.000187508,0.001813381,0.0003515487,0.0002047915,0.0003861456,0.0004654235,0.2084091,0.004276613,0.6257082,0.01896271,0.1389971],"study_design_scores_gemma":[0.00001756666,0.00003336337,0.0002483443,0.00001631712,0.00003776564,0.0001572768,0.00001917855,0.81051,0.000634023,0.1784032,0.009895104,0.00002784289],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005759574,0.0009676288,0.9856995,0.0006978651,0.00008254485,0.00007657433,0.0006543316,0.001043658,0.005018403],"genre_scores_gemma":[0.5493494,0.003569176,0.4092633,0.001407789,0.0007845023,0.001133261,0.004346891,0.0008364758,0.02930925],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006065931,"threshold_uncertainty_score":0.01639086,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02058730128944972,"score_gpt":0.2583534641804145,"score_spread":0.2377661628909648,"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."}}