{"id":"W1592805114","doi":"","title":"Analyzing linguistic data : a practical introduction to statistics using R","year":2008,"lang":"en","type":"book","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2077,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Variety (cybernetics); Computer science; Variation (astronomy); Construct (python library); Range (aeronautics); Visualization; Natural language processing; Measure (data warehouse); Linguistics; Artificial intelligence; Statistical model; Computational statistics; Data science; Data mining; Machine learning; Engineering","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":[0.00769965,0.001955887,0.002001739,0.003701233,0.000842249,0.004002429,0.002889315,0.00230962,0.05917167],"category_scores_gemma":[0.03045329,0.002101705,0.001763541,0.005024508,0.001835347,0.004937917,0.001949587,0.005964647,0.08156703],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009599385,"about_ca_system_score_gemma":0.002127981,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001309381,"about_ca_topic_score_gemma":0.001857521,"domain_scores_codex":[0.9944729,0.0024065,0.0006055127,0.0005106286,0.001917918,0.00008640739],"domain_scores_gemma":[0.9746906,0.0198489,0.00067102,0.001776948,0.002777156,0.0002354246],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001863094,0.00003959185,0.0002674185,0.001330242,0.00008458409,0.0002669269,0.0004819488,0.002408553,0.001136848,0.05373962,0.6551655,0.28506],"study_design_scores_gemma":[0.00001587805,0.00003246352,0.0006770855,0.0005696905,0.00001673884,0.0005497842,0.0001289484,0.003625225,0.0004775641,0.1050539,0.8887912,0.00006158437],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0003626689,0.01870219,0.9054471,0.01036109,0.002624973,0.0004689365,0.005519016,0.01654414,0.03996995],"genre_scores_gemma":[0.003805503,0.01707313,0.9218673,0.004382249,0.002379028,0.001591782,0.003856228,0.007889781,0.03715499],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.05917167,"threshold_uncertainty_score":0.197949,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06438462978172278,"score_gpt":0.3688184923184362,"score_spread":0.3044338625367134,"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."}}