{"id":"W2163814946","doi":"10.1145/1557670.1557681","title":"Keyword query cleaning using hidden Markov models","year":2009,"lang":"en","type":"article","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ontario Institute of Technology","funders":"","keywords":"Computer science; Hidden Markov model; Query optimization; Probabilistic logic; Web query classification; Data mining; Information retrieval; Keyword spotting; Web search query; Artificial intelligence; Search engine","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":[0.00321279,0.0008873787,0.002166094,0.001526935,0.0009381374,0.002355125,0.00257634,0.001717738,0.001642767],"category_scores_gemma":[0.01649788,0.0009979465,0.001677713,0.002794567,0.0008370521,0.004344387,0.001949835,0.001639775,0.00137307],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001444097,"about_ca_system_score_gemma":0.001742694,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00976168,"about_ca_topic_score_gemma":0.007739646,"domain_scores_codex":[0.9961842,0.001197616,0.0003040927,0.0008468953,0.001208278,0.0002588823],"domain_scores_gemma":[0.9893778,0.006915462,0.0006390671,0.001911619,0.0009669325,0.0001892409],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001140368,0.0002594443,0.006599985,0.0006131851,0.0003710988,0.0008371228,0.0009752242,0.5487764,0.02320324,0.04162165,0.008937226,0.366665],"study_design_scores_gemma":[0.00002089996,0.00003903137,0.0002793494,0.000009849026,0.00003222142,0.0001312163,0.00006002086,0.9722988,0.005583444,0.01945412,0.002063001,0.00002807207],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01046668,0.0004866214,0.9851569,0.000233453,0.0000251309,0.00005927047,0.0002437356,0.002983042,0.0003451268],"genre_scores_gemma":[0.3674482,0.0007409338,0.6265253,0.0003660923,0.0001070811,0.0001727626,0.001887587,0.0005416095,0.002210354],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00976168,"threshold_uncertainty_score":0.01940972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0369287846093509,"score_gpt":0.2555249096461961,"score_spread":0.2185961250368452,"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."}}