Feature subset evaluation

A feature selection process mainly consists of two parts feature subset generation and feature subset evaluation. Feature subset generation is a searching procedure. It generates subsets of features for... 

Wrapper methods utilize the prediction ability of some learning machine (SVM in this book) to evaluate the feature subset. Compared with other methods, wrapper method can assure to get a feature subset with higher accuracy by using the specified learning machine (SVM here). The principle of wrapper methods is to minimize... 

In this section we describe the various experimental techniques which have been used to measure aT and we include a critical evaluation of their limitations so as to aid in the intercomparison of the various sets of data. Many experimental groups have contributed to this field, but the aim here is not to discuss every technique in great detail but rather to select a representative subset in order to illustrate the most interesting features. Discussions of the results obtained are given in sections 2.5 and... 

Genetic Algorithm applied to the discussed data set leads to different subsets of selected variables. There are many different versions of GA, depending on the way reproduction, cross-over, etc., are performed. The algorithm used in our study, adapted from Leardi et al. [34,35], is particularly directed towards feature selection. In each GA run, a few subsets with similar responses are selected. Final solutions are then evaluated based on the RMSEP of an independent test set. Results are presented in Table 2 and in Figs. 5 and 6. 

Fuzzy comprehensive evaluation method which based on fuzzy mathematics is a method of quantitative evaluation. According to the features of Huanghua port and characteristics of storage coal, through the selection of evaluation subset, setting of remark set, and the decision of membership degree, establish evaluation system for this port coal. 

Figure 1 illustrates how the genetic algorithm can be used for feature selection. The genetic algorithm is used to generate possible models, and each model is evaluated for its fitness. A model consists of a subset of descriptors drawn from the available pool of descriptors. The fitness function is a... 

---