Feature Selection for Human Grasping Activity Using Pearson's Correlation Techniques
Keywords:
feature selection, human grasping data, Pearson's correlation, PCA-best matching unit (PCA-BMU), sum of movement (SuM)Abstract
The algorithm of feature selection is the collective of search technique to categorize features into their evaluation score. There are many methods to determine the feature extraction in human grasping analysis such as statistical features, PCA-best matching unit (PCA-BMU) and sum of movement (SuM). Feature selection is important in order to increase the classification accuracy by removing redundant features. In analyzing the human grasping data, only the best features are selected in order to make classifying more accurate, less redundant and quickly identifiable, especially for the objects grouping. Pearson's correlation or simply known as the angular separation is capable to measure the similarity of two vectors rather than the distance or the dissimilarity between them. Advantages of the Pearson's correlation are that it is easy to work out and it's easy to be interpreted.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2016 International Journal of Integrated Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Open access licenses
Open Access is by licensing the content with a Creative Commons (CC) license.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.