{"id":"W157143106","doi":"","title":"A Hierarchical EM Approach to Word Segmentation.","year":2001,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Morpheme; Word (group theory); Artificial intelligence; Computer science; Segmentation; Lexicon; Text segmentation; Natural language processing; Character (mathematics); Speech recognition; Minimum description length; Hidden Markov model; Pattern recognition (psychology); 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001659088,0.0009923606,0.001164106,0.001458022,0.0006662875,0.001464913,0.002437754,0.001515252,0.006090284],"category_scores_gemma":[0.005925932,0.001007441,0.00159153,0.001519953,0.001001618,0.00246356,0.001990527,0.001730407,0.004615468],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001155949,"about_ca_system_score_gemma":0.001590759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005048309,"about_ca_topic_score_gemma":0.00860523,"domain_scores_codex":[0.9989064,0.0004322406,0.00006047369,0.0003373272,0.0001868357,0.00007671343],"domain_scores_gemma":[0.9983833,0.0009628677,0.0001073262,0.0002490873,0.0002352755,0.00006210877],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001417644,0.0001141555,0.00165218,0.0003061979,0.0003043321,0.0001797526,0.000447628,0.4661138,0.008302548,0.08409807,0.01409667,0.4242428],"study_design_scores_gemma":[0.00001215616,0.00001618274,0.0002180839,0.00001105225,0.00001471396,0.00004683139,0.00002257453,0.9524277,0.001172316,0.04272674,0.003318369,0.00001332698],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001259432,0.0001040509,0.9967199,0.0001061319,0.00001955645,0.00003094474,0.00009182379,0.0007241245,0.0009440249],"genre_scores_gemma":[0.09285357,0.0002572906,0.8971863,0.0003998051,0.00009909752,0.0003441976,0.001370621,0.0007157007,0.006773411],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006090284,"threshold_uncertainty_score":0.020374,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01821088677380965,"score_gpt":0.2825513774056881,"score_spread":0.2643404906318785,"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."}}