{"id":"W2085380703","doi":"10.1109/tevc.2012.2207120","title":"Evolved Features for DNA Sequence Classification and Their Fitness Landscapes","year":2013,"lang":"en","type":"article","venue":"IEEE Transactions on Evolutionary Computation","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Overfitting; Computer science; Artificial intelligence; Cluster analysis; Random forest; Set (abstract data type); Fitness landscape; Sequence (biology); Feature (linguistics); Genomics; Machine learning; Annotation; Pattern recognition (psychology); Genome; Biology; Artificial neural network; Population","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":[0.001726057,0.0006465732,0.0006849556,0.001647611,0.0004724185,0.001009171,0.0007007629,0.001295225,0.0009815861],"category_scores_gemma":[0.008487149,0.0003053544,0.0006468327,0.000752783,0.0008423157,0.001092962,0.0006469018,0.0009165491,0.0001537054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001130039,"about_ca_system_score_gemma":0.0003379264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001816809,"about_ca_topic_score_gemma":0.001186496,"domain_scores_codex":[0.9994923,0.0001988097,0.00003088876,0.0001026337,0.0001270269,0.00004832804],"domain_scores_gemma":[0.996361,0.002822702,0.0001868579,0.0001886931,0.0003690966,0.00007162322],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001450787,0.0001005241,0.005927912,0.00006748101,0.00007112436,0.0001074027,0.0001076059,0.8943176,0.00555498,0.007641023,0.000627366,0.085332],"study_design_scores_gemma":[0.00001051237,0.00004031864,0.0007670182,0.000006643531,0.000007929441,0.00002832442,0.0000135952,0.9931045,0.0008783242,0.004937842,0.0001976341,0.00000729986],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5801843,0.0007103838,0.4150615,0.0006038268,0.00004474662,0.0001030399,0.0002208497,0.0006602336,0.002411077],"genre_scores_gemma":[0.8660663,0.0001417422,0.131875,0.00008962626,0.00002274175,0.0001937225,0.0003741714,0.0001086715,0.001128138],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001816809,"threshold_uncertainty_score":0.009128392,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02041303771068374,"score_gpt":0.2465370289284259,"score_spread":0.2261239912177422,"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."}}