Main Article Content
Due to digital advancement, the whole world contains the huge amount of data (irrelevant and redundant data). If we use such type of data then the performance of the system degrades. So, Prominent Feature Selection problem is one of the challenging problems in the Machine Learning Area. Principal component analysis (PCA) is the dimensionality reduction (DR) technique in which the original features are transformed from higher dimensional space into lower dimensional space. Though the PCA space has orthogonal principal components (PC), it does not provide a real reduction of dimensionality in terms of the original features (variables), as all features including irrelevant and redundant features are required to define a single PC. It is necessary to remove such type of features by using feature subset selection (FSS) for better generalization performance. The key objective of this paper is to introduce a PCA based Prominent Feature Selection for Random Forest with Fuzzy Logic algorithm (PF-FRF) approach which is able to handle uncertainty classification problem. The PF-FRF is divided into five subparts : Input, PCA, FSS, fuzzification and classification. FSS selects an Prominent feature order which is based on maximum occurrences within the various filter based ranking algorithms. The simulation results are computed and compared by using sequential forward search strategy for clinical datasets. With the results, it is inferred that PF-FRF provides 5.5% improved generalization performance as compared to P-FRF (without feature subset selection).