{"id":"W2067662418","doi":"10.1145/1344411.1344415","title":"Locality-Based pruning methods for web search","year":2008,"lang":"en","type":"article","venue":"ACM Transactions on Information Systems","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Fundação para a Ciência e a Tecnologia; Conselho Nacional de Desenvolvimento Científico e Tecnológico","keywords":"Pruning; Computer science; Locality; Search engine; Simple (philosophy); Reduction (mathematics); Data mining; Artificial intelligence; Phrase; Information retrieval; Machine learning; Mathematics","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":[0.001493684,0.0006104015,0.001143008,0.003196417,0.0008827439,0.001072088,0.001475164,0.0008994366,0.001775872],"category_scores_gemma":[0.009582659,0.0004133847,0.0006276796,0.003596255,0.0008079034,0.002464399,0.001242181,0.0009222531,0.0009059979],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007214796,"about_ca_system_score_gemma":0.000891291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002107868,"about_ca_topic_score_gemma":0.002865003,"domain_scores_codex":[0.9978167,0.0004806798,0.0001525391,0.0002066292,0.001236699,0.0001068476],"domain_scores_gemma":[0.9961197,0.002174235,0.0002988015,0.0007603745,0.0005863649,0.00006049785],"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":[0.0002577046,0.0001472948,0.001699949,0.000457982,0.0001141469,0.0003317846,0.0004467225,0.05292108,0.0289312,0.04168283,0.007419673,0.8655897],"study_design_scores_gemma":[0.0001397844,0.0003412362,0.002904182,0.0001388628,0.0001879873,0.002010667,0.0002258312,0.8187744,0.04444723,0.08799371,0.04273166,0.0001044869],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02359166,0.004265576,0.9671649,0.0002353809,0.00008216256,0.000156405,0.0001540513,0.001590268,0.002759665],"genre_scores_gemma":[0.2257821,0.002380399,0.765236,0.0001788744,0.0002772862,0.0004268116,0.0006296634,0.0003690814,0.004719846],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003196417,"threshold_uncertainty_score":0.007899404,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0624142167311531,"score_gpt":0.3439271494300092,"score_spread":0.2815129326988561,"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."}}