{"id":"W4283450505","doi":"","title":"Sharing HTR datasets with standardized metadata: the HTR-United initiative","year":2022,"lang":"en","type":"preprint","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Research Data Management Practices","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Metadata; Computer science; World Wide Web; Information retrieval; 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":["metaresearch","open_science"],"consensus_categories":[],"category_scores_codex":[0.1167301,0.00123229,0.0017976,0.00751574,0.003700922,0.01367445,0.007715777,0.004017022,0.02272322],"category_scores_gemma":[0.1376169,0.001968513,0.00248021,0.01396378,0.003273368,0.02660078,0.02762327,0.006697773,0.01831502],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003332834,"about_ca_system_score_gemma":0.01973731,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01595488,"about_ca_topic_score_gemma":0.01164407,"domain_scores_codex":[0.9200784,0.03689146,0.00869942,0.006230931,0.02420907,0.003890662],"domain_scores_gemma":[0.7600271,0.03085446,0.006012985,0.1522516,0.03373558,0.0171183],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004965714,0.0002781133,0.001958929,0.0007229382,0.0001857111,0.0002643597,0.001930186,0.001366828,0.004507669,0.09675428,0.7847923,0.1067421],"study_design_scores_gemma":[0.0001209197,0.0001041838,0.001937774,0.0005000108,0.00004628094,0.0001797311,0.0009937881,0.00171127,0.004062186,0.02227299,0.9679283,0.000142553],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02150973,0.005400936,0.5388368,0.0565903,0.01149627,0.00453055,0.146137,0.08383617,0.1316622],"genre_scores_gemma":[0.06866094,0.003078778,0.4076206,0.008856921,0.002428821,0.003340937,0.4414969,0.03414515,0.03037099],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9922842,"threshold_uncertainty_score":0.6173353,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07233787372278151,"score_gpt":0.3178392181969567,"score_spread":0.2455013444741752,"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."}}