{"id":"W2037307735","doi":"10.1016/j.jcss.2007.03.007","title":"Optimal spaced seeds for faster approximate string matching","year":2007,"lang":"en","type":"article","venue":"Journal of Computer and System Sciences","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":false,"ca_institutions":"Simon Fraser University","funders":"","keywords":"String (physics); Mathematics; Matching (statistics); Mathematical optimization; Computer science; Combinatorics; Statistics","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":[0.002842361,0.0001161575,0.0002557275,0.0002268315,0.0003212625,0.000588172,0.0007786054,0.00003714222,4.36088e-7],"category_scores_gemma":[0.00000442192,0.00007630422,0.00007844191,0.0002226782,0.00005621342,0.001187811,0.0002609019,0.00009490303,6.717566e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002502197,"about_ca_system_score_gemma":0.00005558522,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004279294,"about_ca_topic_score_gemma":5.553368e-7,"domain_scores_codex":[0.9985569,0.00003698892,0.0004740321,0.0002164724,0.0004302353,0.0002854208],"domain_scores_gemma":[0.9990218,0.000195188,0.0003678213,0.0001409145,0.0001345765,0.0001396952],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002314774,0.0002797804,0.004757294,0.001140638,0.0002122996,0.0003806823,0.01502451,0.0411601,0.00598198,0.3044351,0.004017284,0.6223788],"study_design_scores_gemma":[0.0009312855,0.0007379218,0.002264638,0.0005724663,0.00001298032,0.001455198,0.001209683,0.9892783,0.0009487065,0.0007464544,0.001574217,0.0002681424],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1280967,0.0002205041,0.870279,0.000127515,0.001024567,0.00009063702,0.000001117219,0.00002312791,0.0001368373],"genre_scores_gemma":[0.5260027,0.000003125984,0.4736226,0.00003255671,0.0003244535,4.291334e-7,1.129307e-7,0.000002518603,0.00001146237],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9481182,"threshold_uncertainty_score":0.5671756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02193825351894405,"score_gpt":0.2730451647751626,"score_spread":0.2511069112562185,"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."}}