{"id":"W2093641143","doi":"10.3115/1073445.1073477","title":"Frequency estimates for statistical word similarity measures","year":2003,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":197,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Coordenação de Aperfeiçoamento de Pessoal de Nível Superior","keywords":"Computer science; Word (group theory); Word lists by frequency; Natural language processing; Similarity (geometry); Artificial intelligence; Context (archaeology); Synonym (taxonomy); Semantic similarity; Set (abstract data type); Word2vec; Information retrieval; Sentence; Linguistics","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":[0.01439873,0.001078008,0.001286129,0.013098,0.0007595656,0.002593651,0.001747386,0.001716024,0.003376941],"category_scores_gemma":[0.1654783,0.0006367235,0.001030136,0.007947293,0.001211253,0.006815277,0.001734604,0.001784346,0.00150911],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001048012,"about_ca_system_score_gemma":0.0007432044,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001088337,"about_ca_topic_score_gemma":0.0009039005,"domain_scores_codex":[0.9829086,0.007392759,0.00154123,0.002133162,0.005728808,0.000295443],"domain_scores_gemma":[0.8117186,0.1594287,0.007500245,0.01259244,0.008221723,0.0005383765],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007217142,0.000437342,0.03669973,0.001518127,0.0007909104,0.0002325312,0.001331701,0.111429,0.01498541,0.1126834,0.007007783,0.7121624],"study_design_scores_gemma":[0.0001856592,0.0006005115,0.02759832,0.0003330548,0.0001860976,0.001393788,0.0007400032,0.7638283,0.01956507,0.1707585,0.01449094,0.0003197967],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03205218,0.001120326,0.9634184,0.0001237823,0.00007420548,0.0002147321,0.0007352756,0.0008311162,0.001429972],"genre_scores_gemma":[0.313215,0.0007771584,0.6808929,0.00009979081,0.0003702739,0.001072166,0.002446176,0.0003404558,0.000786052],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01439873,"threshold_uncertainty_score":0.07614863,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05852592657578515,"score_gpt":0.2991449372864159,"score_spread":0.2406190107106308,"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."}}