{"id":"W4387123803","doi":"10.1109/rew57809.2023.00022","title":"Automatic Domain-Specific Corpora Generation from Wikipedia - A Replication Study","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Baseline (sea); Crawling; Encoder; Artificial intelligence; Natural language processing; Workflow; Replication (statistics); Replicate; Domain (mathematical analysis); Realm; World Wide Web; Information retrieval; Database","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02398724,0.001239039,0.0009303547,0.002240432,0.001486591,0.002205538,0.002420721,0.001564728,0.00292771],"category_scores_gemma":[0.09555209,0.0007619968,0.001396896,0.002027474,0.001507055,0.0036267,0.003082449,0.002116556,0.002430073],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009143762,"about_ca_system_score_gemma":0.00175046,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00449543,"about_ca_topic_score_gemma":0.00522045,"domain_scores_codex":[0.9786309,0.0126152,0.0023254,0.003688085,0.002280241,0.0004601267],"domain_scores_gemma":[0.8802814,0.04337326,0.00260518,0.05443012,0.01825733,0.001052732],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.006596104,0.01365378,0.07432155,0.008314174,0.002083105,0.004061973,0.030081,0.02548366,0.09261643,0.01091217,0.08851674,0.6433593],"study_design_scores_gemma":[0.006503592,0.01471701,0.1148458,0.001718166,0.002958045,0.0063449,0.02231445,0.113076,0.2267038,0.02533745,0.4643613,0.001119305],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8021989,0.00183785,0.1360899,0.00143003,0.001477778,0.009390168,0.0163615,0.01123035,0.01998357],"genre_scores_gemma":[0.6838626,0.0005740481,0.2608918,0.00116102,0.000265053,0.01078983,0.03316772,0.002918325,0.006369735],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9760128,"threshold_uncertainty_score":0.1268581,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0838222573724981,"score_gpt":0.2817979867145097,"score_spread":0.1979757293420116,"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."}}