RETRACTED ARTICLE: Effective feature selection technique in an integrated environment using enhanced principal component analysis

Autor: H. Srimathi, D. Hemavathi
Rok vydání: 2020
Předmět:
Zdroj: Journal of Ambient Intelligence and Humanized Computing. 12:3679-3688
ISSN: 1868-5145
1868-5137
DOI: 10.1007/s12652-019-01647-x
Popis: Dataset have various number of features. Feature extraction plays a crucial job in recognition and extraction of most useful data from the dataset. Appropriate mining method must be performed in order to remove the required information from the dataset. Feature selection helps to train the machine learning algorithm faster as well as reduce the complexity of the model. The hidden patterns are learnt and perceived the unlabeled data in unsupervised learning. Supervised learning alludes to the ability of learning and organizing the signal. This research focus the feature selection by the Enhanced Principal Component Analysis (EPCA) algorithm. This is appropriate for supervised, unsupervised environment. Double measures are associated with EPCA technique as feature selection and the feature extraction. These two measures are used for removing the unnecessary features in the data using dissimilarity matrix calculation. At that point this calculation was contrasted with different calculations by ensuring that the important features were never neglected by EPCA. This EPCA gives better results in supervised, unsupervised environment and it has been tested various numerical and text data in various learning environment.
Databáze: OpenAIRE
Externí odkaz: