Dimension Manipulation Based 360 Degrees to Excavate Clusters in Star Coordinate

Star Coordinate (SC) is one of the visualization techniques that are used for visualizing high dimensional data. SC provides interactive features that can be manipulated to reveal clusters patterns for quick summary and decision. The interactive features offered are dimension arrangement and scaling. However, dimension arrangement feature will be highlighted only. This technique plots the dimension into based 360-degree environment. Unfortunately, finding the most appropriate dimension angle by novice users so that they are aware of the existing clusters is challenging. This paper proposed to assist novice users in manipulating interactive SC features to obtain better clusters visualization for user understanding. The proposed method involved five (5) phases; 1. Calculate the distance between individual attributes against a dependent attribute using Euclidean Distance (ED); 2. Calculate the correlation value between data attributes using Pearson (PR) correlation; 3. Sort the obtained correlation value in ascending order; 4. Reordered data attributes with the positive values to the right and negative values to the left according to the correlation value; 5. The resulting tables are applied to produce the SC based angle 360 degrees. Results achieved by comparing the performance of clusters appearance between dimensions angle in 360 degrees based correlation value and within an equal angle. As a conclusion, the proposed method aware the users of the existing cluster. Thus, assists them for further clusters exploration.