{"id":"W2565143233","doi":"10.82308/24058","title":"Text classification using a hidden Markov model","year":2005,"lang":"en","type":"article","venue":"Open MIND","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Hidden Markov model; Computer science; Artificial intelligence; Categorization; Natural language processing; Naive Bayes classifier; Test set; Machine learning; Bayesian probability; Pattern recognition (psychology); Data mining; Support vector machine","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":[0.001433453,0.0006481193,0.0008869658,0.001420802,0.000631158,0.001230765,0.001475162,0.001239894,0.002496913],"category_scores_gemma":[0.00480981,0.0003755238,0.001098551,0.001184459,0.0004774725,0.00199629,0.0005466169,0.001177964,0.001953888],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001125272,"about_ca_system_score_gemma":0.001009951,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01245833,"about_ca_topic_score_gemma":0.009555318,"domain_scores_codex":[0.9989958,0.0003168589,0.0000789195,0.0003098605,0.0002184095,0.00008007138],"domain_scores_gemma":[0.9972562,0.001996267,0.0001606669,0.0001638864,0.0003629818,0.00006005996],"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":[0.0006303106,0.0004201392,0.008972483,0.0003931301,0.0002398313,0.0004233718,0.0005023355,0.403645,0.01184685,0.02463869,0.008554343,0.5397335],"study_design_scores_gemma":[0.000008750714,0.00002820117,0.0004353583,0.000009741379,0.00001773728,0.00002701573,0.00001155067,0.9935811,0.0006627971,0.004537301,0.000670739,0.000009726971],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03501368,0.0006898452,0.9572354,0.0005995274,0.0001521081,0.0001910524,0.0006661327,0.003503189,0.001949028],"genre_scores_gemma":[0.6791013,0.001011928,0.3089161,0.0004635365,0.0002491584,0.0005517472,0.002048981,0.0001833395,0.007473975],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01245833,"threshold_uncertainty_score":0.02477163,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1071301472310728,"score_gpt":0.3418787696192839,"score_spread":0.2347486223882111,"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."}}