{"id":"W4367310939","doi":"10.1145/3543873.3587671","title":"Gone, Gone, but Not Really, and Gone, But Not forgotten: A Typology of Website Recoverability","year":2023,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Web page; Computer science; Crawling; World Wide Web; Web content; Content analysis; Web crawler; Typology; Content (measure theory); Information retrieval; History; Mathematics","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":[0.00614721,0.0003917954,0.0003350536,0.008271282,0.002586462,0.00373497,0.001388158,0.001187244,0.001541801],"category_scores_gemma":[0.03745092,0.0002952818,0.0006447893,0.004587821,0.008862077,0.007939132,0.004078535,0.001608162,0.0001950642],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001885279,"about_ca_system_score_gemma":0.001111565,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005627452,"about_ca_topic_score_gemma":0.00437481,"domain_scores_codex":[0.992945,0.002508447,0.0005526077,0.0007770984,0.002557051,0.0006596193],"domain_scores_gemma":[0.954055,0.02888808,0.006653038,0.00578722,0.003554746,0.001061931],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003826149,0.0001796362,0.27401,0.0009457644,0.000215567,0.002841653,0.5055291,0.003426281,0.01426111,0.05563189,0.005171311,0.1374052],"study_design_scores_gemma":[0.0000289631,0.0003325377,0.2933046,0.0008613144,0.0001269499,0.004854253,0.5548412,0.01241068,0.00765472,0.05689673,0.06835368,0.0003343438],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9523955,0.0005838737,0.02863618,0.001370305,0.00003057279,0.0001658697,0.0004579727,0.0003787278,0.01598115],"genre_scores_gemma":[0.9931306,0.000155082,0.00508253,0.00008317096,0.0000146451,0.00006655048,0.0002765218,0.0001127272,0.001078109],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008271282,"threshold_uncertainty_score":0.03250992,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02559890688976391,"score_gpt":0.2598745866211452,"score_spread":0.2342756797313813,"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."}}