{"id":"W3098665331","doi":"10.29173/cais1145","title":"Quality Matters: A New Approach for Detecting Quality Problems in Web Archives","year":2020,"lang":"en","type":"article","venue":"Proceedings of the Annual Conference of CAIS / Actes du congrès annuel de l ACSI","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"World Wide Web; Web site; Computer science; Quality (philosophy); Similarity (geometry); Humanities; Library science; Information retrieval; The Internet; Art; Artificial intelligence; Image (mathematics); Philosophy","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001331674,0.000257398,0.0006570346,0.0001574211,0.0001034128,0.002055731,0.003293508,0.0000971444,0.0000038342],"category_scores_gemma":[0.01456173,0.0002075339,0.0002576922,0.0007741734,0.0001962861,0.006006038,0.001003178,0.0002934025,0.000001227949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002450809,"about_ca_system_score_gemma":0.0003208523,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003689553,"about_ca_topic_score_gemma":0.0000250167,"domain_scores_codex":[0.9976207,0.00007311268,0.0008238653,0.0006293788,0.000413974,0.0004389718],"domain_scores_gemma":[0.9926149,0.0004096375,0.001051952,0.0003538662,0.005354994,0.0002146171],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003904146,0.0005800556,0.3482587,0.004206012,0.0004154058,0.000001015555,0.3202463,0.0003311753,0.1950841,0.03900754,0.005325853,0.0861534],"study_design_scores_gemma":[0.01012751,0.001886921,0.2484591,0.002400695,0.0004334627,0.00005504308,0.04445484,0.439964,0.1403823,0.05062256,0.05701012,0.004203341],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.906348,0.00009590763,0.07383268,0.01530857,0.00006118611,0.0006948299,0.000198659,0.0001347862,0.003325347],"genre_scores_gemma":[0.9652402,0.00001523953,0.03387165,0.0006162883,0.00007579802,0.00003631852,0.000005484785,0.00001433383,0.0001247145],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4396329,"threshold_uncertainty_score":0.9989802,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0750485152629651,"score_gpt":0.2913280194931412,"score_spread":0.2162795042301761,"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."}}