{"id":"W7116988931","doi":"10.2139/ssrn.5958498","title":"Privacy Preserving Semantic Search over Encrypted Cloud Email Archives using Bidirectional Encoder Representations from Transformer","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Cryptography and Data Security","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Marriott International (Canada)","funders":"","keywords":"Encryption; Cloud computing; Cryptography; Encoder; Embedding; Confidentiality; Cloud storage; Public-key cryptography; Discrete cosine transform","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","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001287651,0.0004644405,0.0005004063,0.0008138689,0.000792881,0.0007283493,0.002780343,0.0002827101,0.0001145672],"category_scores_gemma":[0.00009638449,0.0004639589,0.0006034049,0.0007943283,0.0001322052,0.0008748475,0.001247638,0.006131635,0.000005966213],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004820129,"about_ca_system_score_gemma":0.00697559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003108363,"about_ca_topic_score_gemma":0.001516248,"domain_scores_codex":[0.9943674,0.0005813016,0.0007691202,0.001077073,0.0008928472,0.002312321],"domain_scores_gemma":[0.9976465,0.0004688932,0.0003070835,0.001219984,0.0001607179,0.0001968164],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004173498,0.001248521,0.02283585,0.0004384003,0.005473679,0.00007685333,0.02093482,0.01875887,0.007822608,0.8699573,0.001016014,0.05101976],"study_design_scores_gemma":[0.0008034302,0.000060404,0.004224101,0.0004159775,0.0001671723,0.0001536338,0.0004695446,0.09866938,0.0005935653,0.8929079,0.0009361646,0.0005987974],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1603961,0.0036558,0.8322734,0.0006427371,0.001302521,0.0002896863,0.0001461626,0.0001360994,0.001157551],"genre_scores_gemma":[0.8831564,0.01907248,0.09349854,0.000249797,0.002669346,0.00006377419,0.0003986428,0.00007997298,0.0008110331],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7387748,"threshold_uncertainty_score":0.9997812,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02447954522190493,"score_gpt":0.3037421129250845,"score_spread":0.2792625677031795,"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."}}