{"id":"W2590383062","doi":"10.1016/j.patcog.2017.02.020","title":"PSI: Patch-based script identification using non-negative matrix factorization","year":2017,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Non-negative matrix factorization; Artificial intelligence; Pattern recognition (psychology); Identification (biology); Normalization (sociology); Representation (politics); Matrix decomposition; Set (abstract data type)","routes":{"ca_aff":true,"ca_fund":true,"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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0004353996,0.0002377746,0.0002109556,0.0002775881,0.0007513686,0.001170553,0.0008343994,0.0001589645,0.00009630238],"category_scores_gemma":[0.0002223542,0.0002587413,0.0001115762,0.0001719232,0.00008227484,0.002283429,0.0001428699,0.0001868913,0.0004287565],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001489323,"about_ca_system_score_gemma":0.00007511953,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000520298,"about_ca_topic_score_gemma":0.00005638445,"domain_scores_codex":[0.9980933,0.0001320271,0.0004545485,0.0005999489,0.0004113941,0.0003087533],"domain_scores_gemma":[0.9976566,0.00006957877,0.000731144,0.0009038053,0.0005313939,0.000107408],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0000134676,0.0001977237,0.0124505,0.0001116426,0.00003386111,0.00001878562,0.0004986859,0.00001021198,0.06597605,0.00003709374,0.000330776,0.9203212],"study_design_scores_gemma":[0.0008973497,0.0000975801,0.03544667,0.0003532008,0.00004418142,0.00001405164,0.00004147175,0.1972732,0.7478638,0.01725466,0.00008650419,0.0006273032],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1251782,0.000006173723,0.8725756,0.0003237956,0.0006889067,0.0004974173,0.0000674066,0.0003080511,0.000354491],"genre_scores_gemma":[0.9658728,0.00001304385,0.03335666,0.0002380174,0.0001775488,0.0001012402,0.0001621303,0.00002866343,0.00004995052],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9196939,"threshold_uncertainty_score":0.9999865,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05510593265417631,"score_gpt":0.3205421891221328,"score_spread":0.2654362564679565,"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."}}