{"id":"W1636186653","doi":"10.48550/arxiv.1012.0663","title":"An Effective Clustering Approach to Web Query Log Anonymization","year":2010,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Database transaction; Cluster analysis; Transaction log; Data mining; Anonymity; Generalization; Information retrieval; Web search query; Transaction data; Search engine; Database; 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":[0.003028335,0.0007070026,0.001043738,0.002331075,0.001822201,0.001696859,0.002307394,0.001269172,0.0009600428],"category_scores_gemma":[0.008842689,0.0003804562,0.001122512,0.003779979,0.001486551,0.003851754,0.002382169,0.001902623,0.0005343299],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001493335,"about_ca_system_score_gemma":0.001712345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002975316,"about_ca_topic_score_gemma":0.003144151,"domain_scores_codex":[0.9955379,0.001766662,0.0002530943,0.0008937787,0.001314779,0.00023373],"domain_scores_gemma":[0.9932938,0.001512563,0.0005805327,0.003569857,0.0008981173,0.0001451626],"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":[0.0003704291,0.0003391212,0.004752289,0.0002046167,0.0002542895,0.0003301167,0.001250793,0.3617058,0.01474635,0.28089,0.0127732,0.322383],"study_design_scores_gemma":[0.00001901005,0.00006178841,0.00129093,0.00001967659,0.00004330083,0.0003575074,0.0002292522,0.8769545,0.006092018,0.106401,0.008477079,0.0000540179],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0151731,0.0002662637,0.9820064,0.0003871937,0.0000427888,0.0001034227,0.0002802597,0.0004210269,0.001319586],"genre_scores_gemma":[0.4958443,0.000717046,0.4965761,0.0003299121,0.0002725907,0.0003513001,0.00146985,0.0001843846,0.004254392],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003028335,"threshold_uncertainty_score":0.01601553,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05374275022465443,"score_gpt":0.209077431378426,"score_spread":0.1553346811537716,"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."}}