{"id":"W4409061687","doi":"10.1007/978-3-031-82362-6_9","title":"PhishCoder: Efficient Extraction of Contextual Information from Phishing Emails","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Phishing; Computer science; Information extraction; World Wide Web; Computer security; Information retrieval; Artificial intelligence; 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":[0.0003593277,0.001633507,0.001418449,0.004182073,0.0007050618,0.001089734,0.0009660743,0.00117616,0.007649916],"category_scores_gemma":[0.001949786,0.0006096817,0.0007523527,0.003064428,0.0003223898,0.001466988,0.001741029,0.0007218507,0.0123088],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003328891,"about_ca_system_score_gemma":0.0008176013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003193134,"about_ca_topic_score_gemma":0.005703032,"domain_scores_codex":[0.9994616,0.00005069827,0.00003569322,0.000160903,0.0002067729,0.00008441187],"domain_scores_gemma":[0.9992265,0.0002423778,0.00008087275,0.0001843453,0.0002025226,0.00006341131],"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":[0.0004706067,0.0002268474,0.004622902,0.0006483913,0.00009334337,0.0005490088,0.0003563755,0.002033388,0.05938327,0.001666129,0.09789594,0.8320538],"study_design_scores_gemma":[0.0002372099,0.001098868,0.04545739,0.0003893628,0.0004615579,0.004278453,0.0009171723,0.424286,0.2799144,0.01995752,0.2226762,0.0003259692],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1559801,0.007128656,0.5802022,0.0008526552,0.0009205436,0.001012198,0.03682071,0.20328,0.01380295],"genre_scores_gemma":[0.2806913,0.002589972,0.6116045,0.0005147014,0.0006540453,0.0006336967,0.07234473,0.004176226,0.02679085],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007649916,"threshold_uncertainty_score":0.02559155,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0117224431044922,"score_gpt":0.2361510860793195,"score_spread":0.2244286429748273,"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."}}