{"id":"W2902584426","doi":"10.1002/pra2.2018.14505501036","title":"What's cached is prologue: Reviewing recent web archives research towards supporting scholarly use","year":2018,"lang":"en","type":"article","venue":"Proceedings of the Association for Information Science and Technology","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Scholarship; Web engineering; Field (mathematics); World Wide Web; Sociotechnical system; Computer science; Data science; Web intelligence; Web development; The Internet; Knowledge management; Political science","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":["metaresearch","scholarly_communication"],"consensus_categories":["scholarly_communication"],"category_scores_codex":[0.007502274,0.00007906186,0.0001585049,0.000997574,0.0009074348,0.002226469,0.001575799,0.000103234,0.000001117519],"category_scores_gemma":[0.01484938,0.00005825337,0.00003658489,0.004701777,0.0005087986,0.02234329,0.0008805003,0.000236064,0.000006820252],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001234578,"about_ca_system_score_gemma":0.0003228218,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007006311,"about_ca_topic_score_gemma":0.000001266642,"domain_scores_codex":[0.9980448,0.0000116279,0.0004437281,0.0002429751,0.0008648094,0.0003919883],"domain_scores_gemma":[0.9938595,0.0001037318,0.0007465773,0.0002272549,0.00501794,0.00004493697],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0000144098,0.00004588137,0.0674173,0.0001454355,0.00006214302,3.629679e-8,0.01468639,4.94843e-7,0.02844309,0.2265497,0.00771277,0.6549224],"study_design_scores_gemma":[0.001193348,0.0007080599,0.01691863,0.0005652572,0.00007388807,0.00001946499,0.01239838,0.06815935,0.3228526,0.05724666,0.5192775,0.0005869178],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8917016,0.0001332046,0.001648841,0.09963936,0.0005611288,0.001084504,0.0000218786,0.0002909853,0.00491848],"genre_scores_gemma":[0.9883584,0.0005590641,0.01028224,0.0005050964,0.00003063629,0.00004896656,0.000001929004,0.000002941571,0.0002107683],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6543354,"threshold_uncertainty_score":0.9988093,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06167223104596187,"score_gpt":0.3423555068436749,"score_spread":0.2806832757977131,"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."}}