When the interpretability is an important issue, then classification methods such as Classification and Regression Trees (CART) might be more attractive, since they are designed to detect the important predictor variables and, for each predictor variable, the critical values which are most relevant for classification.
However, this did not work in all cases and additionally has not been validated as the most clinically relevant classification.
It should be noted that the top ranked features are not necessarily the ones that are individually most relevant to the classification performance since the relevance of features are evaluated in the context of other features, i.e., mutual information among features in terms of their collective discerning power are considered.
In this paper, a new hybrid feature selection scheme is used to determine most relevant features for classification of benign and malignant tumors in breast ultrasound images.
The attribute at the root node, in particular, is considered the most relevant attribute for classification.
We propose an enhanced binary PSO (EPSO) to select a small (near-optimal) subset of informative genes that is most relevant for cancer classification.
To achieve the necessary dimensionality reduction, a watershed segmentation algorithm is then used to group voxels into regional clusters and to identify the most relevant features to classification (group discrimination) [ 7].
Thus the algorithm finds which kernels, and hence which features per data type, are most relevant for the given classification problem.
Choosing the most relevant features to improve the classification accuracy is a challenging problem [6].
