{"id":"W4399254988","doi":"10.1145/3652963.3655042","title":"A Large Scale Study and Classification of VirusTotal Reports on Phishing and Malware URLs","year":2024,"lang":"en","type":"article","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Phishing; Malware; Computer science; Scale (ratio); Computer security; World Wide Web; Information retrieval; The Internet","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":[0.002874295,0.0004253448,0.0004139571,0.01132212,0.000770939,0.001223101,0.0004626356,0.0007642035,0.0006542662],"category_scores_gemma":[0.01632315,0.0002377939,0.0004765012,0.005642853,0.0007846461,0.002260557,0.001294619,0.0008684484,0.0008685903],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000502851,"about_ca_system_score_gemma":0.0004398842,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003760905,"about_ca_topic_score_gemma":0.00566692,"domain_scores_codex":[0.996473,0.0008847928,0.0003954662,0.0008203457,0.0011665,0.0002598767],"domain_scores_gemma":[0.9657974,0.0135532,0.0106308,0.002906053,0.005728315,0.00138422],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001051032,0.0002080007,0.956098,0.0002440962,0.0001531421,0.0004467597,0.003997979,0.0003775143,0.003066242,0.0002524297,0.002878347,0.03217251],"study_design_scores_gemma":[0.000004759605,0.0001545391,0.9883077,0.00008270338,0.00004940159,0.0008450092,0.002373275,0.003470073,0.001499922,0.0001396073,0.003035876,0.00003720219],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9953995,0.0003936051,0.0008777238,0.00007396121,0.00002107094,0.00007376791,0.002339773,0.0001028527,0.0007176533],"genre_scores_gemma":[0.9907415,0.0003203595,0.002141703,0.00006741247,0.00006270518,0.00007637872,0.005823718,0.00004389593,0.000722329],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01132212,"threshold_uncertainty_score":0.01520091,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02001285813462681,"score_gpt":0.266938786538445,"score_spread":0.2469259284038182,"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."}}