Improving feature selection performance using pairwise pre-evaluation

Abstract

BackgroundBiological data such as microarrays contain a huge number of features. Thus, it is necessary to select a small number of novel features to characterize the entire dataset. All combinations of the features subset must be evaluated to produce an ideal feature subset, but this is impossible using currently available computing power. Feature selection or feature subset selection provides a sub-optimal solution within a reasonable amount of time.ResultsIn this study, we propose an improved feature selection method that uses information based on all the pairwise evaluations for a given dataset. We modify the original feature selection algorithms to use pre-evaluation information. The pre-evaluation captures the quality and interactions between two features. The feature subset should be improved by using the top ranking pairs for two features in the selection process.ConclusionsExperimental results demonstrated that the proposed method improved the quality of the feature subset produced by modified feature selection algorithms. The proposed method can be applied to microarray and other high-dimensional data.

Accurate and fast feature selection workflow for high-dimensional omics data

“…Finally, we applied the same FS workflow to gene expression data from normal and prostate tumor tissues ( Datasets 5, 6 and 7 ), and compared them with the results obtained by Li et al . [ 22 ], who used a similar approach on the same datasets (see Table 9 in [ 22 ]). Even though we observe a slight improvement in the classification accuracy in these three datasets ( Table 3 ), the most notable differences were found in the number of features obtained by the final models and in the total runtime, using a similar computational platform.…”

“…Recently, Li et al . have used several gene expression datasets to benchmark different FS algorithms [ 22 ]. From the original microarray datasets, we have selected three of those datasets (GEO accession number: GSE6919), to compare the FS workflow with the results obtained by Li et al .…”

Accurate and Fast feature selection workflow for high-dimensional omics data

“…We have demonstrated the impact of the FS workflow in the classification and/or regression results as well as in the performance of the machine learning algorithm (CPU time and memory). Finally, we applied the same FS workflow to gene expression data from normal and prostate tumor tissues (Datasets 5, 6 and 7), and compared them with the results obtained by Li et al [22], who used a similar approach on the same datasets (see Table 9 in [22]). Even though we observe a slight improvement in the classification accuracy in these three datasets (Table 3), the most notable differences were found in the number of features obtained by the final models and in the total runtime, using a similar computational platform.…”

“…Thus, the results from the comparison reinforce our previous observations and validate the effectiveness of the FS workflow proposed in this manuscript. Table 3: Performance comparison between the proposed approach (X2-PCA-RFE-RF) and the method reported by Li et al [22]. The computer used in the original manuscript was an Intel(R) Core(TM) i5-4690 @ 3.5 GHz CPU, with 16 GB of RAM.…”

Hybrid Support Vector Machine to Preterm Birth Prediction

“…Feature selection is an important issue in the classification model. The reduction of feature is helpful to improve the prediction accuracy and computation time [13]. There is some functions kernel in SVM such as linear, polynomial and RBF.…”