Developing a Filter-Wrapper Feature Selection Method and its Application in Dimension Reduction of Gen Expression

Nowadays, increasing the volume of data and the number of attributes in the dataset has reduced the accuracy of the learning algorithm and the computational complexity. A dimensionality reduction method is a feature selection method, which is done through filtering and wrapping. The wrapper methods are more accurate than filter ones but perform faster and have a less computational burden. With respect to the advantages and disadvantages of the filter and wrapper algorithms, a new hybrid approach is proposed in this study. In the method, all features in the dataset are considered, then the optimal subset of features is selected by combining the feature selection filter algorithms and evaluating their results using the wrapper method. Considering the many diseases and biosystem issues, such as cancer, can be identified and diagnosed by microarray data analysis and considering that there are many features in such datasets, the method proposed in this paper has been evaluated on microarray data related to three types of cancers.  Compared with similar methods, the results show the proposed method benefits from high accuracy in classifying and identifying the affecting factors on cancer.