{"id":"W2129123581","doi":"10.3115/1641976.1641983","title":"Evaluation of several phonetic similarity algorithms on the task of cognate identification","year":2006,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Similarity (geometry); Context (archaeology); Hidden Markov model; Identification (biology); Task (project management); Artificial intelligence; Set (abstract data type); Cognate; Cluster analysis; Natural language processing; Speech recognition; Pattern recognition (psychology); Image (mathematics); Linguistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01063181,0.001940462,0.001999336,0.005301379,0.001606493,0.002100168,0.002538792,0.003365715,0.001525281],"category_scores_gemma":[0.02996798,0.0003795664,0.0009248203,0.0030116,0.0008555969,0.003644646,0.002317194,0.001155716,0.0009671666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001511505,"about_ca_system_score_gemma":0.001901077,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007179587,"about_ca_topic_score_gemma":0.007905433,"domain_scores_codex":[0.990844,0.003578581,0.001084542,0.002144969,0.001870354,0.0004775079],"domain_scores_gemma":[0.9734163,0.01774825,0.0008792107,0.002050682,0.004991803,0.0009137011],"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":[0.003130011,0.001602826,0.03092067,0.0006948307,0.001094194,0.0001699475,0.0004711653,0.1069447,0.009268522,0.003084132,0.003931363,0.8386876],"study_design_scores_gemma":[0.0003716075,0.001930607,0.01393618,0.00006297805,0.0003022886,0.0003593299,0.0006523729,0.9527168,0.02314618,0.004150058,0.002287405,0.00008413165],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8143561,0.005991656,0.1642024,0.0004723881,0.0003294525,0.0005138362,0.0008471738,0.005378193,0.00790873],"genre_scores_gemma":[0.8107142,0.0007194906,0.1829876,0.0001571219,0.000102794,0.0002131759,0.003081481,0.0003111195,0.001713002],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01063181,"threshold_uncertainty_score":0.05622709,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02583546049157253,"score_gpt":0.2935216319955214,"score_spread":0.2676861715039489,"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."}}