{"id":"W4399003852","doi":"10.7910/dvn/x27dse","title":"Text data for comparing the language used in Wikipedia, the DiabeticConnect Forum, and Research Articles","year":2019,"lang":"en","type":"dataset","venue":"Harvard Dataverse","topic":"Wikis in Education and Collaboration","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University; University of British Columbia","funders":"","keywords":"Computer science; Information retrieval; World Wide Web; Library science; Data 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001850037,0.001749679,0.001005695,0.006955922,0.001369548,0.002187668,0.001813669,0.001894536,0.03287151],"category_scores_gemma":[0.01057436,0.0004885205,0.0008596601,0.007463634,0.0005348024,0.001494745,0.002734979,0.001895907,0.05034636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001759055,"about_ca_system_score_gemma":0.00274893,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01431267,"about_ca_topic_score_gemma":0.03182567,"domain_scores_codex":[0.9973111,0.0004301603,0.0004651248,0.0005831196,0.0008834788,0.0003271186],"domain_scores_gemma":[0.9935862,0.001914109,0.0007688584,0.001021472,0.002005702,0.0007036374],"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":[0.0001396155,0.000140912,0.004643486,0.001010244,0.00004409269,0.00006038187,0.0002018336,0.0002724609,0.0005232889,0.0009686378,0.9842386,0.007756526],"study_design_scores_gemma":[0.0002575897,0.00005872036,0.02288656,0.0004215405,0.00005365198,0.0001474128,0.000630363,0.000871064,0.001723968,0.001304724,0.9715685,0.00007584282],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001146145,0.00004588557,0.0001229897,0.00008356739,0.0000411502,0.00004422428,0.9973362,0.0002383091,0.0009416131],"genre_scores_gemma":[0.0009646,0.00002557559,0.0004264154,0.00003082735,0.00001110358,0.0001792834,0.9975061,0.00005370037,0.0008024107],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03287151,"threshold_uncertainty_score":0.1099662,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1286646354725352,"score_gpt":0.4146913308654475,"score_spread":0.2860266953929123,"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."}}