{"id":"W2140060501","doi":"10.1145/1557670.1557680","title":"Query segmentation using conditional random fields","year":2009,"lang":"en","type":"article","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Computer science; Conditional random field; Query optimization; Query expansion; Segmentation; Quality (philosophy); Sargable; Data mining; Field (mathematics); Information retrieval; Web query classification; Relational database; Artificial intelligence; Web search query; 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":[0.003558432,0.0008621759,0.001485497,0.00241777,0.0007100384,0.001097786,0.0018534,0.001317318,0.002239048],"category_scores_gemma":[0.01000164,0.0006173545,0.001494496,0.002794532,0.0009744731,0.003188057,0.0009929725,0.0013208,0.0007014534],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001748521,"about_ca_system_score_gemma":0.00181713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01589428,"about_ca_topic_score_gemma":0.009228011,"domain_scores_codex":[0.9976954,0.000865808,0.0001174005,0.0006025353,0.000483247,0.0002356672],"domain_scores_gemma":[0.9926406,0.005580576,0.000444005,0.0007084579,0.0004849702,0.000141539],"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":[0.00112522,0.0002676902,0.006174789,0.0002685743,0.0001597848,0.0003921079,0.0005211755,0.6150312,0.01316585,0.03758425,0.01257438,0.3127351],"study_design_scores_gemma":[0.00001741759,0.00004486196,0.0004408738,0.000007860903,0.00001523587,0.00008041001,0.00001907543,0.982251,0.001323299,0.01477988,0.001000509,0.00001952693],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01899962,0.0007493331,0.9758671,0.0002578487,0.00003297204,0.00008190677,0.0003250576,0.00296409,0.0007221292],"genre_scores_gemma":[0.6878771,0.0007134589,0.3051723,0.0003877387,0.0002153472,0.0002576408,0.002437505,0.0003957746,0.002543201],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01589428,"threshold_uncertainty_score":0.03160352,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02237004621615805,"score_gpt":0.2745286333615625,"score_spread":0.2521585871454045,"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."}}