{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004886962,0.00007124065,0.0001052564,0.00005678581,0.00007369423,0.0001532592,0.001929417,0.00003594319,0.0001326729],"category_scores_gemma":[0.00005302694,0.00005858377,0.00002269343,0.0002443005,0.00002048039,0.001259321,0.001504886,0.00007937849,0.0003047677],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009377461,"about_ca_system_score_gemma":0.00003572897,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007882718,"about_ca_topic_score_gemma":0.0000478118,"domain_scores_codex":[0.9990102,0.00005378101,0.0001227719,0.0003074046,0.0002146743,0.0002911486],"domain_scores_gemma":[0.9979225,0.0001411984,0.00002303352,0.001772405,0.00002044849,0.0001203962],"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":[0.000002996623,0.0002442116,0.4140649,0.00001215411,0.0002007542,0.00002431736,0.004012604,0.00003362962,0.001058189,0.0364116,0.1257444,0.4181902],"study_design_scores_gemma":[0.0004828057,0.00004302531,0.07371912,0.00004257861,0.00007926107,0.0000156773,0.001440312,0.7455769,0.004337328,0.0004752864,0.1729331,0.000854641],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.142505,0.0003433448,0.8455454,0.001266314,0.00036902,0.00002561835,0.0000420132,0.0001864058,0.009716837],"genre_scores_gemma":[0.7638428,0.00001307817,0.2337232,0.0004026624,0.0003526883,7.284924e-7,0.0001855255,0.000004569391,0.001474831],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7455432,"threshold_uncertainty_score":0.3917274,"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."}}