{"id":"W4404424456","doi":"10.2139/ssrn.5021795","title":"Maximizing Information Gain in Privacy-Aware Active Learning of Email Anomalies","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Personal Information Management and User Behavior","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Information gain; Computer science; Internet privacy; Data mining","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.007740195,0.0002811857,0.0004893392,0.001914087,0.0001434362,0.0007298252,0.001172919,0.0002192569,0.0001284474],"category_scores_gemma":[0.0007111722,0.000230573,0.0003123315,0.0007324557,0.000055317,0.001535784,0.001225708,0.005413004,0.000220845],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001137301,"about_ca_system_score_gemma":0.00220385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009440525,"about_ca_topic_score_gemma":0.0002563341,"domain_scores_codex":[0.99516,0.0002372064,0.001557527,0.0002595671,0.001557924,0.001227715],"domain_scores_gemma":[0.9977249,0.0001892858,0.001294787,0.0003179916,0.0004078431,0.00006519094],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004609487,0.0001225184,0.02354239,0.0003181681,0.0004799098,0.0000221033,0.03479923,0.03855917,0.00007261786,0.1149531,0.001597026,0.7850728],"study_design_scores_gemma":[0.001005499,0.0002594222,0.01084873,0.0006138338,0.0001298955,0.0001017927,0.101476,0.02992725,0.0001709754,0.8472944,0.007487654,0.0006846106],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9790888,0.0009184523,0.0130281,0.001186969,0.0006632629,0.000368491,0.00002244583,0.00005600648,0.004667425],"genre_scores_gemma":[0.9965416,0.001023626,0.0001382168,0.00005451917,0.00009707648,0.00001620766,0.00003623481,0.00001498917,0.00207755],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7843882,"threshold_uncertainty_score":0.9968815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09276002986324035,"score_gpt":0.3839509198760234,"score_spread":0.291190890012783,"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."}}