Parallel Rough Set: Dimensionality Reduction and Feature Discovery of Multi-Dimensional Data in Visualization

Attempt to visualize high dimensional datasets typically encounter over plotting and decline in visual comprehension that makes the knowledge discovery and feature subset analysis difficult. Hence, reshaping the datasets using dimensionality reduction technique is paramount by removing the superfluous attributes to improve visual analytics. In this work, we applied rough set theory as dimensionality reduction and feature selection methods on visualization to facilitate knowledge discovery of multi-dimensional datasets. We provided the case study using real datasets and comparison against other methods to demonstrate the effectiveness of our approach.