{"id":"W2136711891","doi":"10.1109/cibcb.2010.5510426","title":"Nearest neighbor training of side effect machines for sequence classification","year":2010,"lang":"en","type":"article","venue":"","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Computer science; Leverage (statistics); Artificial intelligence; Rand index; Support vector machine; Feature (linguistics); k-nearest neighbors algorithm; Machine learning; Pattern recognition (psychology); Algorithm; Cluster analysis","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":[0.001911582,0.0004613837,0.0009131944,0.0007083326,0.0005149743,0.0005171371,0.001244699,0.00143204,0.002155374],"category_scores_gemma":[0.009217422,0.0003417517,0.0005437313,0.000515501,0.0004879943,0.001058953,0.00046541,0.001235318,0.000746663],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006087594,"about_ca_system_score_gemma":0.0006154059,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002123023,"about_ca_topic_score_gemma":0.002623799,"domain_scores_codex":[0.9987805,0.0005108883,0.00009933108,0.0002531096,0.0002619857,0.00009417599],"domain_scores_gemma":[0.9957348,0.002709475,0.0001879004,0.0005087575,0.0007699417,0.0000891958],"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":[0.0004535079,0.0003030745,0.003641729,0.0001310531,0.00007890674,0.0001195229,0.0001092052,0.57003,0.005642831,0.006287285,0.00266919,0.4105337],"study_design_scores_gemma":[0.000006039189,0.0000418721,0.0002248193,0.000003676228,0.000002683823,0.00001719044,0.000007133624,0.9964167,0.001176552,0.001897445,0.0002019223,0.000004069889],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2091214,0.001180393,0.7817609,0.0004571141,0.0001939808,0.0001731858,0.0001918,0.001614614,0.005306602],"genre_scores_gemma":[0.8404154,0.0001482006,0.1559852,0.0001408475,0.00005529256,0.0001836662,0.0004465445,0.00006522329,0.002559646],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002155374,"threshold_uncertainty_score":0.01010954,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02344163874877428,"score_gpt":0.3139397900067358,"score_spread":0.2904981512579615,"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."}}