{"id":"W2122160657","doi":"10.1007/s10032-005-0013-6","title":"Feature selection for ensembles applied to handwriting recognition","year":2006,"lang":"en","type":"article","venue":"International Journal on Document Analysis and Recognition (IJDAR)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":47,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure","funders":"","keywords":"Artificial intelligence; Random subspace method; Computer science; Feature selection; Pattern recognition (psychology); Boosting (machine learning); Perceptron; Machine learning; Robustness (evolution); Cascading classifiers; Feature (linguistics); Artificial neural network; Hidden Markov model; Classifier (UML)","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":[0.001978799,0.0007636899,0.001615685,0.0009129412,0.0006004188,0.0007659358,0.0009000547,0.0006599234,0.00165887],"category_scores_gemma":[0.003757126,0.0004110705,0.001086742,0.001063364,0.0002531757,0.0008212356,0.0009932446,0.001126802,0.0006356151],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003345346,"about_ca_system_score_gemma":0.0006576818,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003069062,"about_ca_topic_score_gemma":0.003805996,"domain_scores_codex":[0.9992477,0.0002540123,0.00008360092,0.0001374479,0.000193431,0.00008387587],"domain_scores_gemma":[0.9977908,0.001102224,0.0001026138,0.0002575888,0.0006593676,0.00008737555],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003503788,0.0001941943,0.002524455,0.00008942215,0.0002668937,0.00008995665,0.00007840437,0.1719244,0.01479146,0.001897092,0.003186971,0.8046066],"study_design_scores_gemma":[0.000009417725,0.00007484534,0.001136092,0.00000812043,0.00003837263,0.00003541869,0.00001258057,0.9927301,0.003700427,0.001579511,0.0006669623,0.000008147719],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05965969,0.0008336931,0.9372509,0.0001226101,0.0001708858,0.00007558809,0.0001547397,0.001056564,0.0006753748],"genre_scores_gemma":[0.7178832,0.0005438796,0.2758777,0.00009719721,0.0002073768,0.0002510023,0.0009006371,0.0002326851,0.004006259],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003069062,"threshold_uncertainty_score":0.01046503,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01284765546056195,"score_gpt":0.2665302875074339,"score_spread":0.253682632046872,"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."}}