{"id":"W2039447541","doi":"10.1145/564376.564448","title":"The impact of corpus size on question answering performance","year":2002,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Terabyte; Question answering; Computer science; Information retrieval; World Wide Web; Natural language processing","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":[0.01604118,0.001266864,0.002080804,0.002302729,0.002031347,0.004490243,0.001954387,0.002123216,0.005413017],"category_scores_gemma":[0.1336818,0.001226227,0.0008176861,0.003299821,0.001506777,0.009438394,0.003028657,0.001842038,0.00316816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001257331,"about_ca_system_score_gemma":0.001937355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006996242,"about_ca_topic_score_gemma":0.007310003,"domain_scores_codex":[0.9848387,0.007107618,0.002057667,0.002479654,0.002858903,0.0006574888],"domain_scores_gemma":[0.8332128,0.142969,0.001974744,0.008808692,0.01106298,0.001971799],"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":[0.01593118,0.002993579,0.05905309,0.004737078,0.001162855,0.001932565,0.004540185,0.0594523,0.1594008,0.004260576,0.0978894,0.5886463],"study_design_scores_gemma":[0.002837961,0.01055252,0.08905075,0.0007058725,0.002627199,0.004648715,0.004963093,0.5001194,0.3007023,0.0155989,0.06743614,0.000757053],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9362507,0.008690783,0.01787249,0.004187755,0.00106907,0.0004252201,0.005305243,0.01232332,0.01387537],"genre_scores_gemma":[0.9433951,0.002666305,0.02718621,0.0009258395,0.000705206,0.0006459939,0.01666844,0.002241198,0.005565571],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01604118,"threshold_uncertainty_score":0.08483487,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01666226182163508,"score_gpt":0.2518497277000022,"score_spread":0.2351874658783671,"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."}}