{"id":"W2885004725","doi":"10.1101/376848","title":"Extracting human data from published figures: implications for data science and bioethics","year":2018,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Information retrieval; Data science; Graphics; Raw data; Data sharing; Reading (process); Linguistics; Programming language","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":["metaresearch","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.1014361,0.0008138089,0.001079092,0.009661129,0.002640535,0.02150967,0.003661561,0.003581563,0.01807134],"category_scores_gemma":[0.5164301,0.0009798085,0.00127474,0.01824024,0.009236045,0.01681023,0.006780273,0.005745206,0.01086423],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003187735,"about_ca_system_score_gemma":0.008529394,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001680224,"about_ca_topic_score_gemma":0.001575448,"domain_scores_codex":[0.9229007,0.05244688,0.006825054,0.004779603,0.01222074,0.0008270172],"domain_scores_gemma":[0.3245252,0.4907931,0.03403903,0.09667713,0.0504552,0.003510275],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004926948,0.0002310432,0.0173957,0.002566506,0.0002641933,0.001321907,0.007990506,0.003473201,0.002330325,0.2632648,0.2670563,0.4336126],"study_design_scores_gemma":[0.00006439689,0.00005547765,0.006084028,0.003728062,0.0000967684,0.0009220588,0.00428608,0.007761587,0.005495146,0.5741152,0.3972235,0.00016766],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02208457,0.01562451,0.5942969,0.2821145,0.01540402,0.001161398,0.01832194,0.00610865,0.04488348],"genre_scores_gemma":[0.2036234,0.01488587,0.7243644,0.01630292,0.008879525,0.001460238,0.01593209,0.003668625,0.01088289],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9964184,"threshold_uncertainty_score":0.536452,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1546460821422862,"score_gpt":0.352421805782515,"score_spread":0.1977757236402288,"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."}}