{"id":"W4286331442","doi":"10.1109/saner53432.2022.00116","title":"Phishing Kits Source Code Similarity Distribution: A Case Study","year":2022,"lang":"en","type":"article","venue":"2022 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa; IBM (Canada); Polytechnique Montréal","funders":"","keywords":"Phishing; Computer science; Identifier; Similarity (geometry); Source code; Code (set theory); World Wide Web; Credit card; Identification (biology); Computer security; Information retrieval; Artificial intelligence; The Internet; Programming language; Payment","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"],"consensus_categories":[],"category_scores_codex":[0.000706345,0.0002412532,0.0002897511,0.0005239872,0.000757068,0.0003416378,0.0006918229,0.00006478976,0.0001781878],"category_scores_gemma":[0.0001157032,0.0002620089,0.0001760857,0.001342895,0.00003499991,0.0004274081,0.0003518859,0.0006001268,0.00000642499],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004322825,"about_ca_system_score_gemma":0.00006922286,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008099476,"about_ca_topic_score_gemma":0.000301372,"domain_scores_codex":[0.9976228,0.0001799715,0.0003955575,0.0006825342,0.0008538574,0.0002652545],"domain_scores_gemma":[0.9988985,0.0001074453,0.0001822108,0.0004613561,0.000206734,0.0001437186],"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":[0.000168085,0.001617684,0.07727651,0.0000500225,0.003860363,0.002638065,0.005771745,0.8390437,0.0009015846,0.04404686,0.003667716,0.02095766],"study_design_scores_gemma":[0.0005545574,0.0002991916,0.009008239,0.00001306299,0.0002172228,0.0006912049,0.001351808,0.9837291,0.00003467237,0.0004026553,0.003264018,0.0004342552],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2271079,0.00008395352,0.7707137,0.0004470693,0.0009750599,0.00014881,0.0001000781,0.0003560824,0.00006724818],"genre_scores_gemma":[0.9983791,0.00001795315,0.0008010247,0.000104292,0.0001520153,0.0000725409,0.0001004895,0.00001297673,0.0003596171],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7712712,"threshold_uncertainty_score":0.9999832,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03310186504630544,"score_gpt":0.2763223032600144,"score_spread":0.243220438213709,"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."}}