{"id":"W1958345963","doi":"10.1007/978-3-540-88192-6_26","title":"Using Data Mining Methods to Predict Personally Identifiable Information in Emails","year":2008,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary; Research and Productivity Council","funders":"","keywords":"Computer science; Private information retrieval; Information extraction; Data mining; Quality (philosophy); Information retrieval; Data science; Computer security","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002609501,0.0003805078,0.0004265595,0.001620439,0.0002879351,0.0009728844,0.004501799,0.0002768455,0.000008344771],"category_scores_gemma":[0.0003856356,0.0003892104,0.000059629,0.001253357,0.0001924407,0.004153815,0.002539105,0.0006155334,0.00002533558],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004154967,"about_ca_system_score_gemma":0.0006770334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002033665,"about_ca_topic_score_gemma":0.0001513595,"domain_scores_codex":[0.9965122,0.00007992124,0.0006476542,0.001209712,0.0009771573,0.0005733597],"domain_scores_gemma":[0.9971119,0.0004127771,0.0002901301,0.001830917,0.0001910837,0.0001631527],"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":[0.000008517,0.00001315685,0.000130057,0.00004358283,0.000008742688,0.00004113434,0.006092625,0.1336299,0.0002645139,0.000250683,0.0001568889,0.8593602],"study_design_scores_gemma":[0.0001636887,0.00007505878,0.0002383291,0.000449411,0.00000533597,0.0001430061,8.158352e-7,0.9908233,0.0005585671,0.003231952,0.003857669,0.0004529076],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0005595914,0.0001955453,0.9949786,0.0002129534,0.002656811,0.0003745407,0.00001185118,0.0001204465,0.0008896693],"genre_scores_gemma":[0.01087664,0.00002166464,0.9876785,0.001004433,0.000303561,0.000004652675,0.00001587976,0.00001741418,0.0000772092],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8589072,"threshold_uncertainty_score":0.999856,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07542289935797529,"score_gpt":0.3271868826411642,"score_spread":0.2517639832831889,"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."}}