An Enhancement of Efficient Clustering High Dimensional Data Analysis Using QDA – UFP Method

  • U. Indumathi, Dr. S. J. Sathish Aaron Joseph

Abstract

In the fast moving digitalized world, the high dimensional data clustering process is the most critical and challenging task. The clustering process helps  to analyze different groups or clusters of similar data objects. These objects are denoted on the point of multidimensional space. The similar data objects are to be identified on the dataset with the help of the distance evaluation. In olden days, traditional clustering algorithms are performed which gives a result like the process of input dataset attempt is too long, slow and difficult for separation of the irrelevant dimensional data object. The research paper proposed a novel technique like enhancement of efficient clustering high dimensional data analysis using QDA – UFP (Quadratic Discriminant Analysis – Univariant Filtering Process) method. In  this paper, the web log or malicious and such as eliminating irrelevant features and eliminating redundant features. Then the features are to be classified using Quadratic discriminant analysis (QDA). The QDA method is more flexible for the covariance matrix which tends to perfect fit to better dataset and also prior knowledge of an individual class. The classified data are processed to the feature selection process. This process is done by Univariant Filtering Process (UFP). This method helps to rank individual features based on specific or predicted criteria. The performance analysis gives high efficiency, effective extraction of features and reducing featured separation error.

Published
2021-07-21
How to Cite
U. Indumathi, Dr. S. J. Sathish Aaron Joseph. (2021). An Enhancement of Efficient Clustering High Dimensional Data Analysis Using QDA – UFP Method. Design Engineering, 2733 - 2751. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/2837
Section
Articles