{"id":"W2140217726","doi":"10.1007/s10032-005-0147-6","title":"Retrieving poorly degraded OCR documents","year":2005,"lang":"en","type":"article","venue":"International Journal on Document Analysis and Recognition (IJDAR)","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure; Université de Montréal; Concordia University","funders":"","keywords":"Computer science; Information retrieval; Precision and recall; Query expansion; Relevance (law); Software; Vector space model; Optical character recognition; Artificial intelligence; Pattern recognition (psychology); Selection (genetic algorithm); Error detection and correction; Document retrieval; Data mining; Natural language processing; Image (mathematics); Algorithm","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":[],"consensus_categories":[],"category_scores_codex":[0.0006145021,0.002406373,0.001891677,0.003890507,0.001034107,0.002114205,0.0007586548,0.002078113,0.01128447],"category_scores_gemma":[0.004239692,0.0004781543,0.0009214867,0.002199588,0.0004579694,0.001703165,0.0008982038,0.001027696,0.009462752],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006847548,"about_ca_system_score_gemma":0.0008230206,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004339493,"about_ca_topic_score_gemma":0.004891699,"domain_scores_codex":[0.9989086,0.00009180759,0.000109297,0.0001652195,0.0005782543,0.0001467563],"domain_scores_gemma":[0.9974585,0.0005581348,0.0002006392,0.0005270564,0.001108982,0.0001466682],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002554022,0.0002740898,0.001946835,0.001030736,0.000154144,0.00208104,0.0001479304,0.005042237,0.5464801,0.0004149548,0.01706991,0.4228039],"study_design_scores_gemma":[0.0001569093,0.001024499,0.0158676,0.0001819898,0.0007033783,0.009102704,0.0006942363,0.1501358,0.7797054,0.00141419,0.04086686,0.0001465992],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7647986,0.01325423,0.1679645,0.001659233,0.001822791,0.0008705674,0.006833723,0.01967346,0.02312295],"genre_scores_gemma":[0.7256222,0.005369714,0.2097605,0.0009609122,0.001127968,0.0001752384,0.0150389,0.002092665,0.03985192],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01128447,"threshold_uncertainty_score":0.0377503,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01535983961365943,"score_gpt":0.3048018974636144,"score_spread":0.2894420578499549,"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."}}