{"id":"W2124618303","doi":"","title":"Substring-Based Transliteration","year":2007,"lang":"en","type":"article","venue":"Meeting of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Transliteration; Substring; Computer science; Margin (machine learning); Artificial intelligence; Word (group theory); Natural language processing; Machine translation; Speech recognition; Programming language; Machine learning; Data structure; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001559742,0.00007827007,0.0001029536,0.00008023988,0.0001817752,0.00006934439,0.0004611382,0.000069058,2.966945e-7],"category_scores_gemma":[0.005592148,0.00006698395,0.00009981616,0.0003080182,0.00001507392,0.000039277,0.00003635957,0.00009306314,5.667156e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001647872,"about_ca_system_score_gemma":0.0000920859,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006651142,"about_ca_topic_score_gemma":0.000006094109,"domain_scores_codex":[0.9988784,0.0000380027,0.0003499077,0.0001443305,0.0004333016,0.0001560965],"domain_scores_gemma":[0.9968214,0.00124031,0.0005063826,0.0001406427,0.00126878,0.00002246633],"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":[0.00003124128,0.0001176552,0.02060572,0.0001353034,0.00004419144,8.324939e-7,0.0006276225,0.1231834,0.0009057717,0.8503111,0.0008605451,0.003176628],"study_design_scores_gemma":[0.0008609492,0.00007698693,0.00763366,0.0001894939,0.00004153016,9.210525e-7,0.000009947464,0.5274749,0.06982344,0.3864295,0.007157406,0.0003012496],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001962826,0.00008958159,0.995899,0.0003936715,0.0006113058,0.0001824885,0.00001247981,0.0001530553,0.000695547],"genre_scores_gemma":[0.5837101,7.615142e-8,0.4160012,0.0001167449,0.0001091209,0.00000223083,0.00001129261,0.000004575466,0.00004457874],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5817474,"threshold_uncertainty_score":0.6694728,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01112075728909654,"score_gpt":0.2755006221297466,"score_spread":0.2643798648406501,"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."}}