Hierarchical Multi-classification with Predictive Clustering Trees in Functional Genomics

摘要: This paper investigates how predictive clustering trees can be used to predict gene function in the genome of yeast Saccharomyces cerevisiae. We consider MIPS FunCat classification scheme, which each is annotated with one or more classes selected from a given functional class hierarchy. setting presents two important challenges machine learning: (1) instance labeled set instead just class, and (2) are structured hierarchy; ideally learning algorithm should also take this hierarchical information into account. Predictive generalize decision applied wide range prediction tasks by plugging suitable distance metric. define an appropriate metric for multi-classification present experiments evaluating approach on number data sets that available yeast.