{"id":"W2740290295","doi":"10.1145/3097570","title":"Warcbase","year":2017,"lang":"en","type":"article","venue":"Journal on Computing and Cultural Heritage","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; World Wide Web; SPARK (programming language); Scalability; Big data; Process (computing); Data science; Web modeling; Analytics; Filter (signal processing); Resource (disambiguation); Web service; 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":[],"consensus_categories":[],"category_scores_codex":[0.002585717,0.001554437,0.001014821,0.004040709,0.002220044,0.01020691,0.004675845,0.002323923,0.189291],"category_scores_gemma":[0.01270431,0.001205599,0.001428624,0.004957642,0.001149668,0.01104315,0.007784723,0.003512985,0.1833314],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001336405,"about_ca_system_score_gemma":0.003546704,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006669295,"about_ca_topic_score_gemma":0.008675506,"domain_scores_codex":[0.9979258,0.0002484124,0.0001861343,0.000462512,0.0009435234,0.0002335964],"domain_scores_gemma":[0.9939216,0.001113816,0.0003088443,0.002554408,0.001467308,0.0006339383],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004219402,0.0001078733,0.001155538,0.0008713371,0.00005611884,0.0003603244,0.001036806,0.0005595397,0.003815475,0.02266934,0.824981,0.1439648],"study_design_scores_gemma":[0.0000335894,0.00001814693,0.0004993012,0.0001245381,0.00001202398,0.0002281554,0.0002117689,0.001272464,0.002284297,0.006809942,0.9884658,0.00004003822],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.007703133,0.003161983,0.1277897,0.00325709,0.002121629,0.001316684,0.1182886,0.3744821,0.361879],"genre_scores_gemma":[0.04033303,0.003768141,0.1730191,0.002614278,0.0004954087,0.001981154,0.2822434,0.1545845,0.340961],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.189291,"threshold_uncertainty_score":0,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03107071811273536,"score_gpt":0.3015826552369414,"score_spread":0.2705119371242061,"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."}}