{"id":"W4297099963","doi":"10.1109/dsc54232.2022.9888835","title":"Malicious and Benign URL Dataset Generation Using Character-Level LSTM Models","year":2022,"lang":"en","type":"article","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Computer science; Blacklist; Classifier (UML); Artificial intelligence; Machine learning; Character (mathematics); Similarity (geometry); Data mining; Information retrieval; Computer security","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":[0.001018811,0.001117798,0.0004059685,0.001588391,0.0003406576,0.0005612118,0.001045174,0.001140207,0.001603313],"category_scores_gemma":[0.005430934,0.0002707638,0.0009721132,0.0008662365,0.0004212761,0.0008690163,0.0008224886,0.0009464559,0.001055548],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008371617,"about_ca_system_score_gemma":0.0004573931,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003614855,"about_ca_topic_score_gemma":0.003472875,"domain_scores_codex":[0.9993062,0.0001758796,0.00009075455,0.0001961319,0.0001619758,0.00006898174],"domain_scores_gemma":[0.9976116,0.0008326556,0.0002133588,0.0005991358,0.0006459811,0.00009721997],"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":[0.001840047,0.002098968,0.02526562,0.001201083,0.0003002055,0.003391549,0.0005232482,0.2239612,0.05474813,0.004226001,0.05221143,0.6302326],"study_design_scores_gemma":[0.00008134484,0.00053235,0.00811497,0.0000477985,0.00005131074,0.0005658634,0.0001532345,0.9450514,0.03592901,0.002410584,0.007001311,0.00006082316],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8073564,0.0007505831,0.132676,0.0009272327,0.0009889742,0.001268218,0.03235361,0.01780914,0.005869789],"genre_scores_gemma":[0.8300946,0.0002273868,0.1135316,0.0001952569,0.00009983139,0.0007595312,0.05199466,0.0002682162,0.00282888],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003614855,"threshold_uncertainty_score":0.007187605,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1701835766062939,"score_gpt":0.2668449957888401,"score_spread":0.09666141918254628,"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."}}