A Set of Complexity Measures Designed for Applying Meta-Learning to Instance Selection
Abstract— A Set of Complexity Measures Designed for Applying Meta-Learning to Instance Selection. Some authors have approached the instance selection problem from a meta-learning perspective. In their work, they try to ﬁnd relationships between the performance of some methods from this ﬁeld and the values of some data-complexity measures, with the aim of < Final Year Projects 2016 > the best performing method given a data set, using only the values of the measures computed on this data. Nevertheless, most of the data-complexity measures existing in the literature were not conceived for this purpose and the feasibility of their use in this ﬁeld is yet to be determined. In this paper, we revise the deﬁnition of some measures that we presented in a previous work, that were designed for meta-learning based instance selection.