{"id":"W2147293034","doi":"10.1145/2348283.2348327","title":"Mining query subtopics from search log data","year":2012,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":65,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Web search query; Information retrieval; Ranking (information retrieval); Cluster analysis; Web query classification; Search engine; Leverage (statistics); Data mining; Artificial intelligence","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":[0.001574862,0.001743722,0.001608239,0.01242097,0.0009117637,0.001346564,0.001109283,0.001026755,0.0006121373],"category_scores_gemma":[0.008716805,0.0004018724,0.001827065,0.007518508,0.0005638376,0.002419779,0.001337046,0.00101287,0.0008979856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009438854,"about_ca_system_score_gemma":0.001684619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01619411,"about_ca_topic_score_gemma":0.02374642,"domain_scores_codex":[0.997203,0.0004446112,0.0003604622,0.0006172497,0.001064936,0.0003098988],"domain_scores_gemma":[0.9934781,0.00316849,0.0007368273,0.000688947,0.001614439,0.0003130892],"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":[0.002480882,0.001199951,0.1925577,0.002852805,0.0005884253,0.002397171,0.003016622,0.03579283,0.1396458,0.004315716,0.02773596,0.5874162],"study_design_scores_gemma":[0.0001354591,0.0008225378,0.1919647,0.0002373808,0.0004510568,0.002746,0.003207405,0.7142742,0.05000475,0.01211362,0.02378086,0.0002619752],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8021984,0.008368642,0.1519096,0.0007052484,0.0001209532,0.001007747,0.02495792,0.006583222,0.004148254],"genre_scores_gemma":[0.8446642,0.001475233,0.1086218,0.0001819829,0.0001412735,0.0003630951,0.04244083,0.0002359909,0.001875571],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01619411,"threshold_uncertainty_score":0.03219968,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1367392867182668,"score_gpt":0.3206255663048285,"score_spread":0.1838862795865617,"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."}}