{"id":"W4406043290","doi":"10.18280/ijsse.140613","title":"Phishing Detection Using Random Forest-Based Weighted Bootstrap Sampling and LASSO+ Feature Selection","year":2024,"lang":"en","type":"article","venue":"International Journal of Safety and Security Engineering","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Random forest; Feature selection; Lasso (programming language); Computer science; Selection (genetic algorithm); Sampling (signal processing); Feature (linguistics); Artificial intelligence; Statistics; Data mining; Machine learning; Pattern recognition (psychology); Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004750034,0.0001099253,0.0001306044,0.0003437367,0.00008650052,0.0004702578,0.0001586462,0.00008576823,0.000001808731],"category_scores_gemma":[0.00007060054,0.0001048194,0.00007002938,0.00019978,0.00001179823,0.0009862956,0.00003571763,0.0004006025,2.215373e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000120842,"about_ca_system_score_gemma":0.0000389857,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002720227,"about_ca_topic_score_gemma":0.00001994505,"domain_scores_codex":[0.9991955,0.00002426196,0.0002380628,0.0001525451,0.0002775261,0.0001121581],"domain_scores_gemma":[0.999469,0.0001856481,0.00008917542,0.00004316252,0.0001467021,0.000066349],"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":[0.001295061,0.0001190364,0.004712681,0.00059355,0.001282522,0.0003096406,0.006132491,0.6067251,0.1672336,0.01906428,0.00004324144,0.1924888],"study_design_scores_gemma":[0.0005843384,0.00005425625,0.002057168,0.0003087674,0.00002054846,0.0007674184,0.00001096651,0.9883294,0.003844129,0.001027738,0.002885104,0.000110201],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2712705,0.0008034983,0.7258499,0.0004626506,0.00150737,0.00003362944,0.000001824165,0.00006167001,0.000008970977],"genre_scores_gemma":[0.9892739,0.0001600325,0.01001442,0.00003053062,0.0005076324,5.4823e-7,0.000001096356,0.000009244545,0.000002608545],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7180034,"threshold_uncertainty_score":0.4534706,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01137422995268639,"score_gpt":0.2443478693589702,"score_spread":0.2329736394062839,"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."}}