{"id":"W2610342131","doi":"10.12688/f1000research.11389.2","title":"PubRunner: A light-weight framework for updating text mining results","year":2017,"lang":"en","type":"preprint","venue":"F1000Research","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canada's Michael Smith Genome Sciences Centre; University of British Columbia","funders":"U.S. National Library of Medicine; National Institutes of Health","keywords":"Computer science; Upload; Workflow; Word2vec; Biomedical text mining; Data science; Field (mathematics); Domain (mathematical analysis); Information retrieval; World Wide Web; Text mining; Data mining; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"not_applicable","genre":"software","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"not_applicable","genre":"software","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04593455,0.005101026,0.00350367,0.02519496,0.004096102,0.01430774,0.009368242,0.004213797,0.02458116],"category_scores_gemma":[0.1324079,0.004654617,0.005550288,0.01741358,0.002928188,0.02846147,0.01352028,0.005745089,0.02733035],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002772233,"about_ca_system_score_gemma":0.008165462,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008580081,"about_ca_topic_score_gemma":0.01514865,"domain_scores_codex":[0.9794862,0.005085807,0.004496268,0.004193583,0.006005807,0.0007323496],"domain_scores_gemma":[0.9226106,0.03712682,0.005403844,0.02072153,0.01131014,0.002827004],"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":[0.001449368,0.0004754152,0.006008897,0.003744162,0.000661293,0.001097561,0.002820329,0.009564828,0.007923806,0.06160478,0.325367,0.5792826],"study_design_scores_gemma":[0.0005690469,0.0002147922,0.002344671,0.001243759,0.0004013291,0.0008878314,0.0007287951,0.1415099,0.02190618,0.2185964,0.6109844,0.0006128532],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.00156371,0.0006695904,0.693276,0.001290763,0.0005735154,0.0008215384,0.01565781,0.2832116,0.002935406],"genre_scores_gemma":[0.0126113,0.0006694063,0.9235989,0.0005652319,0.0002924671,0.0009605762,0.03223226,0.02570628,0.003363472],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04593455,"threshold_uncertainty_score":0.242928,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07276717202123249,"score_gpt":0.3943506542953737,"score_spread":0.3215834822741412,"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."}}