| Autor: |
Khan, Iliyas Karim, Daud, Hanita Binti, Sokkalingam, Rajalingam, Zainuddin, Nooraini Binti, Abdussamad, Naheed, Noor, Iqbal, Mudassar |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3123 Issue 1, p1-7, 7p |
| Abstrakt: |
In this study the new statistical method kernelized rank order (KROD) numerical solution is analyzed to convert non spherical data to spherical form using different samples of data. Non-spherical data can affect the performance of popular algorithms like k-means clustering. The distance between data points in the non-spherical data set is calculated using the KROD method, which uses a rank order distance equation and a Gaussian kernel equation. Each data point is then given a weight based on this distance to convert it into a spherical coordinate. In this numerical solution, pair wise similarity between the data points are weighted using the Gaussian kernel, equation and the actual distance is calculated using the rank order distance (ROD). The combination of the two above methods gives global and local structure of spherical shape with clear visualization. The current study numerically solved the KROD method by applying different samples of data and checking the performance of KROD numerical method to check the effectiveness of spherical visualization. The numerical solution by using differ sample of datasets shows by increasing sample size the KROD statistical model accurately converted non spherical data to spherical data. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
