Volume : VII, Issue : XI, November - 2018

The influence of K proximity selection on some Manifold Learning

Han Baojin

Abstract :

 The selection of K–nearest points in manifold learning has always been an important factor affecting the extraction of some manifold learning features, and how to reasonably select K–nearest points has always been obtained in repeated experiments. This paper uses evolutionary algorithm to find out the different popular learning conditions for different K–nearest points. The experimental results show that the optimal K value in the data set improves the classification accuracy and makes the quality of feature vectors after feature extraction better. 

Keywords :

Article: Download PDF   DOI : 10.36106/ijsr  

Cite This Article:

The influence of K proximity selection on some Manifold Learning , Han Baojin , INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH : Volume-7|Issue-11| November-2018

Number of Downloads : 137

References :