Finally, 'feature selection' was conducted to investigate whether we can still achieve good prediction performance with only a subset of features.
In the latter instance, affected children may present with only a subset of features early in the course of the disease, further complicating the diagnostic process.
To keep the local app database small and portable, we populated it with a subset of feature types that would be most used (Table 1).
Finally, R(f i, C| M s ) (R4-Correlation) is employed as a conditional independence test for identifying redundant features with respect to a subset of features (Corollary 1).
Feature selection (or preprocessing) consists of detecting the relevant features and discarding the irrelevant ones, with the goal of obtaining a subset of features that describe the given problem properly with a minimum degradation of performance.
The proposed method can make use of different representations simultaneously (i.e., multiview learning) to obtain a better prediction performance than using a single feature representation (i.e., single-view learning) or a subset of features, with the advantage of automatic feature selections.
By utilising only structural feature information in our descriptors (structural fingerprints/keys) the descriptor generation of a fragment results in a subset of features with regards to the parent structure.
The ultimate goal of the proposed model is to select a subset of features with accurate classification performance.
After identifying a subset of features with consistent changes across time and sample sets, logistic regression was used to model treatment status (D0 versus M1).
To alleviate this drawback, a recently reported method for feature selection, called the Local Hyperplane-based RELIEfeature feature weighting scheme, can be used to select a subset of features with high prognostic values [ 47- 49].
