{"id":"W4389509223","doi":"10.1002/spe.3296","title":"Parsing millions of URLs per second","year":2023,"lang":"en","type":"article","venue":"Software Practice and Experience","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université TÉLUQ; Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Parsing; Node (physics); Programming language; Parser combinator; Operating system; Artificial intelligence; Physics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001920079,0.00149784,0.001008072,0.005424144,0.0007295409,0.00212115,0.001266908,0.001282226,0.01360274],"category_scores_gemma":[0.01590993,0.001029536,0.0007131936,0.003962989,0.0004645246,0.002592425,0.00133267,0.00125122,0.01421665],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008601479,"about_ca_system_score_gemma":0.001081268,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007518547,"about_ca_topic_score_gemma":0.004302392,"domain_scores_codex":[0.9963669,0.0003398881,0.0003826517,0.0008968044,0.001769174,0.0002446072],"domain_scores_gemma":[0.9900442,0.003081344,0.0005903522,0.001942913,0.004052649,0.0002884779],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000718866,0.0002300264,0.02151231,0.0005862932,0.0001803825,0.0007382947,0.000731327,0.009979249,0.0364897,0.007264175,0.3245692,0.5970002],"study_design_scores_gemma":[0.0001632919,0.0002558735,0.04034897,0.0003044798,0.0001611578,0.001756338,0.0006854638,0.4727606,0.1539777,0.03124284,0.2980189,0.0003242993],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"software","genre_gemma":"methods","genre_scores_codex":[0.1136985,0.001692761,0.308805,0.001532925,0.00132852,0.0005690275,0.04976307,0.503231,0.01937922],"genre_scores_gemma":[0.4130311,0.0009564856,0.4599541,0.0008988876,0.0004776176,0.000526642,0.07844322,0.02181126,0.02390065],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01360274,"threshold_uncertainty_score":0.0455057,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02164362833989121,"score_gpt":0.2929179877798292,"score_spread":0.2712743594399381,"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."}}