{"id":"W4386471798","doi":"10.1186/s42400-023-00183-8","title":"Use of subword tokenization for domain generation algorithm classification","year":2023,"lang":"en","type":"article","venue":"Cybersecurity","topic":"Topic Modeling","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Lexical analysis; Computer science; Artificial intelligence; Word (group theory); Convolutional neural network; Feature (linguistics); Domain (mathematical analysis); Scheme (mathematics); Machine learning; Natural language processing; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.0003081815,0.00006390514,0.00008663138,0.00009626549,0.00008033848,0.00007880297,0.0002045572,0.00005960586,0.000002394338],"category_scores_gemma":[0.00007590337,0.00006852383,0.00004262501,0.0004004311,0.00001712301,0.0006449401,0.00005939542,0.00004038373,0.000009697661],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003555993,"about_ca_system_score_gemma":0.00004081688,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003206329,"about_ca_topic_score_gemma":0.00001937366,"domain_scores_codex":[0.9992108,0.00004354857,0.0002092876,0.0002453356,0.0001569951,0.0001339723],"domain_scores_gemma":[0.9993188,0.00006845529,0.00009545865,0.0003559063,0.0001273176,0.00003409553],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003742356,0.00007856331,0.0007755861,0.00004456939,0.00002075252,0.000001896068,0.004390651,0.002611987,0.02545504,0.5295334,0.006339676,0.4307441],"study_design_scores_gemma":[0.000135157,0.00001571462,0.001746956,0.000004940457,0.000003090413,6.538746e-7,0.00003864954,0.9842003,0.003530646,0.006669486,0.003578489,0.00007593607],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1391781,0.000009134146,0.8594543,0.0005809588,0.0003769757,0.0002190246,0.000008527932,0.0001423534,0.00003064495],"genre_scores_gemma":[0.6771687,0.00001542582,0.3223816,0.00006779196,0.00015773,0.00003980974,0.00009728156,0.000007536213,0.00006406956],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9815883,"threshold_uncertainty_score":0.2794321,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1179084249451668,"score_gpt":0.2921493607814226,"score_spread":0.1742409358362558,"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."}}