Аннотация
Biological sequence analysis presents interesting
challenges for machine learning. Using one of the most
important current problems -- the recognition of
functional target sites for microRNA molecules -- as an
example, we show how joining multiple genetic
programming classifiers improves accuracy and stability
tremendously. When moving from single classifiers to
bagging and boosting with cross validation and
parameter optimisation, you require more computing
power. We use a special-purpose search processor for
fitness evaluation, which renders boosted genetic
programming practical for our purposes.
Пользователи данного ресурса
Пожалуйста,
войдите в систему, чтобы принять участие в дискуссии (добавить собственные рецензию, или комментарий)