Titlebook: Recent Advances in Ensembles for Feature Selection; Verónica Bolón-Canedo,Amparo Alonso-Betanzos Book 2018 Springer International Publishi

脊椎动物

Combination of Outputs,h a final decision. Therefore, a crucial point when designing an ensemble method is to choose an appropriate method for combining the different weak outputs. There are several methods in the literature to solve this issue, and they are grouped according to whether the outputs are classification pred

Addictive

Evaluation of Ensembles for Feature Selection,n, almost universal measures of accuracy, there are two other measures that should be taken into account to quantify the success of an ensemble approach: diversity and stability. In both cases, the relation between the three measures has been studied relatively well in the field of classification en

古董

Other Ensemble Approaches,essfully used. First, in Sect. ., we introduce a very brief review of the different application fields in which ensembles have been applied, together with basic levels that are used to produce different ensemble designs, and a sample taxonomy. Then, in Sect. . basic ideas in ensemble classification

使残废

让步

Software Tools,re is an important number of feature selection and ensemble learning methods already implemented and available in different platforms, so it is useful to know them before coding our own ensembles. Section . comments on the methods available in different popular software tools, such as Matlab, Weka,

Ischemia

cauda-equina

Basic Concepts,valuate the performance of a classifier, whilst in Sect. . different approaches to divide the training set are discussed. Finally, Sect. . gives some recommendations on statistical tests adequate to compare several models and in Sect. . the reader can find some database repositories.

dendrites

Applications of Ensembles Versus Traditional Approaches: Experimental Results,thod for each problem at hand, as it is usually very dependent on the characteristics of the datasets. The adequacy of using an ensemble of filters instead of a single filter is demonstrated on both synthetic and real data, including the challenging scenario of DNA microarray classification.

江湖郎中

Software Tools, to know them before coding our own ensembles. Section . comments on the methods available in different popular software tools, such as Matlab, Weka, R, scikit-learn, or more recent and sophisticated platforms for parallel learning. Then, Sect. . gives some examples of code in Matlab.

frozen-shoulder