TY - GEN
T1 - Feature selection for multi-purpose predictive models
T2 - 11th International Conference on Parallel Problem Solving from Nature
AU - Reynolds, Alan P.
AU - Corne, David W.
AU - Chantler, Michael J.
PY - 2010
Y1 - 2010
N2 - The target of machine learning is a predictive model that performs well on unseen data. Often, such a model has multiple intended uses, related to different points in the tradeoff between (e.g.) sensitivity and specificity. Moreover, when feature selection is required, different feature subsets will suit different target performance characteristics. Given a feature selection task with such multiple distinct requirements, one is in fact faced with a very-many-objective optimization task, whose target is a Pareto surface of feature subsets, each specialized for (e.g.) a different sensitivity/specificity tradeoff profile. We argue that this view has many advantages. We motivate, develop and test such an approach. We show that it can be achieved successfully using a dominance-based multiobjective algorithm, despite an arbitrarily large number of objectives. © 2010 Springer-Verlag.
AB - The target of machine learning is a predictive model that performs well on unseen data. Often, such a model has multiple intended uses, related to different points in the tradeoff between (e.g.) sensitivity and specificity. Moreover, when feature selection is required, different feature subsets will suit different target performance characteristics. Given a feature selection task with such multiple distinct requirements, one is in fact faced with a very-many-objective optimization task, whose target is a Pareto surface of feature subsets, each specialized for (e.g.) a different sensitivity/specificity tradeoff profile. We argue that this view has many advantages. We motivate, develop and test such an approach. We show that it can be achieved successfully using a dominance-based multiobjective algorithm, despite an arbitrarily large number of objectives. © 2010 Springer-Verlag.
UR - http://www.scopus.com/inward/record.url?scp=78149262263&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-15844-5_39
DO - 10.1007/978-3-642-15844-5_39
M3 - Conference contribution
SN - 3642158439
SN - 9783642158438
VL - 6238 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 384
EP - 393
BT - Parallel Problem Solving from Nature, PPSN XI - 11th International Conference, Proceedings
Y2 - 11 September 2010 through 15 September 2010
ER -