{"id":"W3012200654","doi":"10.19195/2084-5065.45.2","title":"How to answer the question of error margin in forensic signature examination with a Bayesian approach?","year":2017,"lang":"en","type":"article","venue":"Nowa Kodyfikacja Prawa Karnego","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Trois-Rivières","funders":"","keywords":"Margin (machine learning); Signature (topology); Computer science; Bayesian probability; Handwriting; False accusation; Certainty; Forensic examination; Forensic science; Artificial intelligence; Machine learning; Psychology; Mathematics; Engineering; Social psychology; Forensic engineering","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001019508,0.0002530399,0.0002898154,0.0003133746,0.0002402974,0.0005315997,0.001567646,0.0001922244,0.000005809015],"category_scores_gemma":[0.000301514,0.0001760402,0.00006456558,0.0004198807,0.0001759165,0.001276243,0.0002837529,0.0003791107,0.000007439843],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000790269,"about_ca_system_score_gemma":0.0001433173,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001389824,"about_ca_topic_score_gemma":0.0005082251,"domain_scores_codex":[0.9980354,0.000263146,0.0002701303,0.0005740338,0.0005196175,0.0003376564],"domain_scores_gemma":[0.997716,0.00008440309,0.0003703298,0.001403527,0.0003210839,0.0001047128],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00007981981,0.0005071276,0.005364157,0.0001600474,0.00004400772,0.00003630962,0.00331891,0.00004368864,0.005254506,0.07296131,0.003737016,0.9084931],"study_design_scores_gemma":[0.005424852,0.003849072,0.6145439,0.002940331,0.0001541086,0.0003924939,0.00237415,0.07965589,0.2033885,0.05389362,0.02972307,0.003659938],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03405707,0.00009956098,0.942656,0.00699484,0.0001956478,0.001544266,0.00000767727,0.0002549797,0.01418992],"genre_scores_gemma":[0.8699937,0.000007428006,0.1285927,0.0003124742,0.00007308105,0.0002137579,0.0000093214,0.00002214335,0.0007753297],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9048331,"threshold_uncertainty_score":0.717871,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02138269685861599,"score_gpt":0.2725000644968411,"score_spread":0.2511173676382251,"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."}}