{"id":"W6893615861","doi":"10.5281/zenodo.3827902","title":"Russian Wikipedia Dump","year":2020,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Wikis in Education and Collaboration","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Optech (Canada)","funders":"","keywords":"The Internet; Government (linguistics); Table (database); Work (physics)","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":[0.0003811525,0.001225608,0.0007769124,0.01040893,0.001543924,0.00169002,0.0006165192,0.0005832993,0.08098391],"category_scores_gemma":[0.00290774,0.0004886676,0.0007673629,0.008603578,0.000370751,0.001589681,0.002070954,0.001009842,0.1038966],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006463684,"about_ca_system_score_gemma":0.002247358,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02093323,"about_ca_topic_score_gemma":0.02728742,"domain_scores_codex":[0.9993081,0.00008245348,0.0001025911,0.0001961433,0.0002177512,0.00009295858],"domain_scores_gemma":[0.9982256,0.0003698572,0.0001421709,0.0002944748,0.0008319101,0.0001358912],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001741973,0.00003819547,0.001466888,0.001756728,0.00006336581,0.000264001,0.0003643406,0.0004192103,0.002201245,0.002001002,0.9582007,0.03305011],"study_design_scores_gemma":[0.00002726788,0.00001001081,0.004403592,0.0001697271,0.00003437286,0.000145522,0.0002322522,0.0002690086,0.001448182,0.0006760377,0.9925671,0.00001690542],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.002922065,0.0003245318,0.001178574,0.0002006638,0.0003689627,0.0000784748,0.9685283,0.002487545,0.02391072],"genre_scores_gemma":[0.005232233,0.0005650286,0.002898484,0.00007667364,0.00005880746,0.0001658286,0.9764193,0.001239058,0.01334452],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.08098391,"threshold_uncertainty_score":0.2709181,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05563039899712723,"score_gpt":0.295894021400576,"score_spread":0.2402636224034487,"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."}}