{"id":"W7154241600","doi":"10.5281/zenodo.19561442","title":"Unlocking Web Histories: Leveraging LLMs and RAG to Transform Discovery in Web Archives","year":2025,"lang":"","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Transparency (behavior); Stewardship (theology); Pipeline (software); Semantic Web; Digital preservation; Digital library; Cultural heritage","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":["metaepi_narrow","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001094651,0.0003022701,0.0003824909,0.00136762,0.002444999,0.003809073,0.00223147,0.00007632088,0.000564098],"category_scores_gemma":[0.0005878999,0.000340478,0.00009403113,0.002684629,0.000270285,0.001230637,0.00300697,0.0005128433,0.0003748448],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002947009,"about_ca_system_score_gemma":0.00004795453,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005378322,"about_ca_topic_score_gemma":0.000008448402,"domain_scores_codex":[0.9966493,0.0004998088,0.0005769295,0.001120119,0.0004447807,0.000709081],"domain_scores_gemma":[0.998489,0.0001043988,0.0001110805,0.000860827,0.0001704739,0.000264284],"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":[0.0001522224,0.0002853907,0.0001063675,0.0003534346,0.0001694864,0.00007628382,0.03232614,0.0007842483,0.009321632,0.01827238,0.02693327,0.9112191],"study_design_scores_gemma":[0.0009319915,0.000192003,0.0005695939,0.0005275814,0.0000431257,0.00003242729,0.001887309,0.06573263,0.0003508636,0.0003674225,0.9289405,0.0004245988],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.139853,0.002058715,0.5670026,0.0233175,0.001013201,0.001506886,0.0007536284,0.00134852,0.2631459],"genre_scores_gemma":[0.9923685,0.0005447445,0.001056228,0.0003537544,0.00008802591,8.301065e-8,0.000207889,0.0004641064,0.004916706],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9107946,"threshold_uncertainty_score":0.9999048,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02313008818105575,"score_gpt":0.2389026225362874,"score_spread":0.2157725343552316,"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."}}