{"id":"W6981196323","doi":"","title":"Duplicate Detection for Quality Assurance of Document Image Collections: Paper - iPRES 2012 - Digital Curation Institute, iSchool, Toronto","year":2012,"lang":"en","type":"article","venue":"Phaidra (Universität Wien)","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Workflow; Robustness (evolution); Digital image; Image file formats; Document image processing; Quality assurance; Image processing; Automatic image annotation","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0002647724,0.000186168,0.00021945,0.0001323455,0.0003072538,0.0002388244,0.0003879017,0.00008328661,0.00003260539],"category_scores_gemma":[0.0001700118,0.0002026568,0.0001444098,0.000553284,0.0001183612,0.878441,0.0001511623,0.00008118412,0.00003958153],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004008098,"about_ca_system_score_gemma":0.00006752063,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003185856,"about_ca_topic_score_gemma":0.0006395071,"domain_scores_codex":[0.9985861,0.00004229711,0.0002946849,0.0003578569,0.0003473591,0.000371747],"domain_scores_gemma":[0.9986966,0.0001377508,0.0002707227,0.000466042,0.0002556621,0.0001732237],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001282766,0.00025418,0.000007581279,0.00007301665,0.00008879614,0.00000100708,0.0009149092,0.00005301442,0.01534754,0.880748,0.0001807077,0.102203],"study_design_scores_gemma":[0.005084786,0.001280556,0.0009542308,0.0001462895,0.0001766445,0.00009477772,0.001561225,0.006230372,0.01321705,0.6771062,0.2924494,0.001698477],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002863754,0.0002512609,0.9836239,0.0002513508,0.001874542,0.0005687621,0.00002906961,0.0002026507,0.01033466],"genre_scores_gemma":[0.989904,0.00002687598,0.007647731,0.00004104636,0.000163123,0.00004942695,0.0000147439,0.00001283005,0.002140249],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9870402,"threshold_uncertainty_score":0.8264104,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01450686990541291,"score_gpt":0.2576323233897625,"score_spread":0.2431254534843496,"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."}}