{"id":"W848342684","doi":"","title":"PaddyWaC: A Minimally-Supervised Web-Corpus of Hiberno-English","year":2011,"lang":"en","type":"article","venue":"Research Portal (Queen's University Belfast)","topic":"Digital Communication and Language","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Irish; Computer science; Variety (cybernetics); Bootstrapping (finance); Artificial intelligence; Domain (mathematical analysis); Varieties of English; Natural language processing; World Wide Web; Linguistics; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007465291,0.000167516,0.0002481905,0.0005351698,0.0002317532,0.00008557212,0.003006832,0.000114689,0.0005516291],"category_scores_gemma":[0.0002236946,0.0001790701,0.0001589553,0.001258347,0.00037366,0.001180131,0.001767566,0.0004717697,0.0001088769],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007432468,"about_ca_system_score_gemma":0.0004027465,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001004811,"about_ca_topic_score_gemma":0.00007220093,"domain_scores_codex":[0.9975221,0.0003661735,0.0002453536,0.000444962,0.0008516475,0.0005697869],"domain_scores_gemma":[0.99713,0.0002219856,0.0001109994,0.001422374,0.0007620839,0.0003525074],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003305266,0.002055464,0.01099062,0.0001694372,0.0002477732,0.002156442,0.01663799,0.000003102839,0.001704206,0.9295922,0.01644995,0.0196623],"study_design_scores_gemma":[0.01008157,0.004119861,0.09734695,0.0006173496,0.00009428551,0.00006344033,0.0411123,0.002787319,0.02664416,0.007370775,0.8066284,0.003133604],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.2415121,0.00003712456,0.0005329006,0.0005127136,0.00007273071,0.0003364434,0.00004015937,0.0002657883,0.7566901],"genre_scores_gemma":[0.9838318,0.00009621452,0.005564305,0.00004595121,0.00001718847,0.00000154654,0.00001333954,0.00001406173,0.0104156],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9222214,"threshold_uncertainty_score":0.7302266,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03669335032176523,"score_gpt":0.2615932642133681,"score_spread":0.2248999138916029,"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."}}