{"id":"W2145050101","doi":"10.1109/cnsr.2007.5","title":"A Novel Approach for Frequent Phrase Mining in Web Search Engine Query Streams","year":2007,"lang":"en","type":"article","venue":"","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick","funders":"Atlantic Canada Opportunities Agency","keywords":"Computer science; Phrase; Set (abstract data type); Data stream mining; Data mining; Information retrieval; Recommender system; Phrase search; Data stream; Web query classification; Web search query; Search engine; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002931794,0.0009556494,0.001398011,0.003528502,0.0007843183,0.002212971,0.001915506,0.001086316,0.001181168],"category_scores_gemma":[0.01126304,0.0005267065,0.001421014,0.00431564,0.0006763,0.004163788,0.001294406,0.001521511,0.0008819969],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006021905,"about_ca_system_score_gemma":0.001166205,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001495713,"about_ca_topic_score_gemma":0.001492727,"domain_scores_codex":[0.9963508,0.0009132443,0.0004274974,0.0007925654,0.001389431,0.0001265458],"domain_scores_gemma":[0.9948334,0.002690738,0.0005381696,0.0007049847,0.001105612,0.0001270454],"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":[0.0005473814,0.0003947633,0.008114827,0.0008444312,0.0002747753,0.0006295678,0.0008828079,0.03768753,0.02787679,0.06165538,0.006964511,0.8541273],"study_design_scores_gemma":[0.00007151407,0.0002759579,0.001849101,0.00006607854,0.0001124643,0.001464505,0.0002439798,0.9210138,0.01240061,0.04552531,0.01688781,0.0000890013],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003564347,0.0002152614,0.995275,0.00009940123,0.00002652607,0.00009168219,0.0001467851,0.0003048944,0.0002760598],"genre_scores_gemma":[0.05770892,0.00043195,0.9398963,0.0001245634,0.0001413956,0.000231071,0.0006767698,0.00006444626,0.000724636],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003528502,"threshold_uncertainty_score":0.01550496,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04000222621762329,"score_gpt":0.2943553656368082,"score_spread":0.2543531394191849,"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."}}