{"id":"W4321487091","doi":"10.1016/j.comnet.2023.109648","title":"Deep learning for encrypted traffic classification in the face of data drift: An empirical study","year":2023,"lang":"en","type":"article","venue":"Computer Networks","topic":"Internet Traffic Analysis and Secure E-voting","field":"Computer Science","cited_by":47,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Encryption; Traffic classification; Granularity; Artificial intelligence; Data mining; Machine learning; Deep learning; Quality of service; Computer network","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":[],"consensus_categories":[],"category_scores_codex":[0.001891402,0.000142607,0.0002507401,0.0001432362,0.000134446,0.0001883903,0.002625177,0.0000784242,0.000001680401],"category_scores_gemma":[0.00002472069,0.0001072355,0.00006969444,0.00108486,0.00003113699,0.0003584603,0.0004092285,0.0003118909,0.000005676212],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001493361,"about_ca_system_score_gemma":0.00002015442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003198653,"about_ca_topic_score_gemma":0.0001118216,"domain_scores_codex":[0.9979066,0.0004565892,0.000449009,0.0006054179,0.00026983,0.0003125408],"domain_scores_gemma":[0.9984748,0.0005437838,0.0001576662,0.0007105513,0.00007071622,0.00004250888],"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":[0.000005288312,0.0001749441,0.001186805,0.000004372003,0.00002534505,0.000005627318,0.009449884,0.8471434,4.845638e-7,0.001320351,0.0009222035,0.1397613],"study_design_scores_gemma":[0.0003017068,0.0002477232,0.02148444,0.00001397289,0.00001539325,0.000002104655,0.001205163,0.9761495,1.21223e-7,0.000005939664,0.0004583824,0.0001155772],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2162302,0.00003935728,0.7829089,0.0001936129,0.0001898933,0.0003089401,3.608031e-7,0.0001201014,0.000008568349],"genre_scores_gemma":[0.9932128,0.000006028059,0.006233782,0.0001148556,0.0002802793,0.00002095654,0.0001102159,0.00001064741,0.00001046523],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7769825,"threshold_uncertainty_score":0.4878275,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08433534082062956,"score_gpt":0.338235020470347,"score_spread":0.2538996796497174,"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."}}