{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006569357,0.0009261337,0.0009666667,0.01583819,0.001417252,0.005114969,0.001570461,0.001917272,0.002354335],"category_scores_gemma":[0.03749436,0.0005887642,0.001186392,0.008640713,0.001951392,0.007566549,0.003486443,0.001694145,0.0008028788],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001801267,"about_ca_system_score_gemma":0.001287871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004573129,"about_ca_topic_score_gemma":0.00358278,"domain_scores_codex":[0.9914824,0.001361851,0.00111972,0.00122204,0.004415483,0.0003984995],"domain_scores_gemma":[0.9686846,0.009889632,0.00767125,0.003091478,0.009721373,0.0009416919],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0008500866,0.0004781726,0.3136153,0.001054066,0.0005167792,0.001195368,0.008804529,0.008482277,0.02825267,0.02443295,0.00678063,0.6055372],"study_design_scores_gemma":[0.000183261,0.001801384,0.4014491,0.0004920358,0.0009855357,0.007056443,0.01684609,0.3909302,0.06554047,0.06163862,0.05254409,0.0005326958],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2988957,0.001607124,0.677345,0.0009685133,0.0002074718,0.0009624959,0.001166222,0.004110491,0.01473694],"genre_scores_gemma":[0.7254095,0.0004389924,0.2693983,0.0001576061,0.0001530809,0.0003133393,0.000691905,0.0002716825,0.003165587],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01583819,"threshold_uncertainty_score":0.03474247,"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."}}