{"id":"W2082347500","doi":"10.1098/rsta.2013.0138","title":"Large-scale detection of repetitions","year":2014,"lang":"en","type":"article","venue":"Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"String (physics); Prefix; Computation; Suffix tree; Factorization; Suffix; Generalized suffix tree; Mathematics; Fraction (chemistry); Combinatorics; Trie; Computer science; Algorithm; Data structure","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":[0.000257966,0.00008951371,0.0001796701,0.00001683479,0.0002252112,0.00002537764,0.0004183435,0.00004215931,0.000003965442],"category_scores_gemma":[0.00004097781,0.00005574253,0.0001997717,0.0003249315,0.0002741317,0.0001617253,0.00007000418,0.0001540164,0.000001261458],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000601928,"about_ca_system_score_gemma":0.000008473445,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000336469,"about_ca_topic_score_gemma":1.544822e-7,"domain_scores_codex":[0.9991629,0.00002331087,0.0001768469,0.0001942974,0.0002950701,0.0001476049],"domain_scores_gemma":[0.9993715,0.0002399507,0.00005050898,0.0002437116,0.00003074241,0.00006362419],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004235184,0.0009455336,0.00001044041,0.0004361545,0.0000513892,9.287468e-8,0.00203194,0.04680274,0.0177841,0.9265618,0.00001107033,0.005360554],"study_design_scores_gemma":[0.00007746929,0.00006840606,0.0001842028,0.00005211632,0.00001235317,0.000001422924,0.00001634083,0.8211377,0.003306062,0.1750505,0.00003343083,0.00005998564],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03900392,0.00001375378,0.9594073,0.001162364,0.00006442349,0.00005764152,0.000006035291,0.00004378427,0.0002407433],"genre_scores_gemma":[0.9699268,0.000002773104,0.02998211,0.0000196431,0.00004996,0.000006818585,1.248151e-7,0.000003163751,0.000008605955],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9309229,"threshold_uncertainty_score":0.2273114,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01044085314286231,"score_gpt":0.2177297521475103,"score_spread":0.207288899004648,"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."}}