{"id":"W4380482005","doi":"10.6000/1929-4409.2020.09.298","title":"Russian Size Adjectives Analyses by Corpus Linguistics Methods","year":2022,"lang":"en","type":"article","venue":"International Journal of Criminology and Sociology","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Kazan Federal University","keywords":"Adjective; Noun; Linguistics; Mathematics; Numeral system; Noun phrase; Parametric statistics; Semantics (computer science); Characterization (materials science); Natural language processing; Statistics; Computer science; Philosophy; Physics; Arithmetic","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002481976,0.0004695562,0.0004113186,0.009666632,0.001913843,0.002142999,0.000475623,0.0003033199,0.0070849],"category_scores_gemma":[0.007380847,0.0003435608,0.0004789259,0.007654221,0.001239856,0.001780043,0.001528884,0.0006657512,0.0016789],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001504332,"about_ca_system_score_gemma":0.001674624,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005467397,"about_ca_topic_score_gemma":0.00759862,"domain_scores_codex":[0.9969862,0.001466222,0.0003826176,0.0006007488,0.0004690381,0.00009521403],"domain_scores_gemma":[0.9924074,0.004842731,0.0007302779,0.0006012304,0.001338898,0.00007958739],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0004909945,0.000242759,0.06310829,0.006005782,0.0002636325,0.002410616,0.1494801,0.002812618,0.04746842,0.1601207,0.05049904,0.517097],"study_design_scores_gemma":[0.00006818074,0.000213855,0.1759019,0.0009649735,0.0004119785,0.002591419,0.06191314,0.01880738,0.02314249,0.03376113,0.6820403,0.0001832214],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6519463,0.005232428,0.1780259,0.001467677,0.0004292013,0.00142792,0.03110214,0.002248331,0.1281201],"genre_scores_gemma":[0.8162495,0.002015693,0.1423948,0.0001857475,0.0001531881,0.002265309,0.01998531,0.0007211125,0.01602939],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009666632,"threshold_uncertainty_score":0.02370137,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0708498737276486,"score_gpt":0.4333488744759437,"score_spread":0.3624990007482951,"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."}}