{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00107302,0.0002326851,0.0005335212,0.0002931347,0.0001525199,0.0001131488,0.0009317719,0.0001577373,0.00007793601],"category_scores_gemma":[0.0003266126,0.0002025703,0.000149802,0.00105951,0.0001884739,0.0004708395,0.0009089477,0.0002009242,0.0002750545],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003296115,"about_ca_system_score_gemma":0.0001215538,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002209532,"about_ca_topic_score_gemma":0.0005282672,"domain_scores_codex":[0.9976169,0.000155894,0.0005501839,0.0008665475,0.0003132134,0.0004972684],"domain_scores_gemma":[0.9976959,0.0004896093,0.0001875003,0.001313345,0.0001470503,0.0001665513],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0005858847,0.001256862,0.1360932,0.0008942395,0.001610493,0.0003717502,0.004205725,0.0005382575,0.19609,0.188412,0.1060823,0.3638593],"study_design_scores_gemma":[0.003457011,0.001434978,0.6511229,0.0001478536,0.0003190978,0.0001484334,0.001410838,0.2399444,0.06519397,0.009327154,0.025271,0.002222386],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9584773,0.00004153524,0.02721494,0.004886747,0.000314089,0.000159107,0.0001976406,0.0004317058,0.008276911],"genre_scores_gemma":[0.9716156,0.000162895,0.01871799,0.000867656,0.00006437492,0.0000151011,0.00005366357,0.00001652517,0.008486252],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5150297,"threshold_uncertainty_score":0.8260579,"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."}}