Advertisement
Published in:
2021 | OriginalPaper | Chapter
Data Compression and Visualization Using PCA and T-SNE
Authors : Jyoti Pareek, Joel Jacob
Published in: Advances in Information Communication Technology and Computing
Publisher: Springer Singapore
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
This paper examines two commonly used data dimensionality reduction techniques, namely, PCA and T-SNE. PCA was founded in 1933 and T-SNE in 2008, both are fundamentally different techniques. PCA focuses heavily on linear algebra while T-SNE is a probabilistic technique. The goal is to apply these algorithms on MNIST dataset and to see how they practically work and what conclusions we could draw from their application. The objective is to reduce the dimension of the data while retaining most of the information. We perform both these techniques and make a comparison between them by observing the results. We note the behavior of the reduced components obtained from both techniques, by visualizing it in 2-dimensional space. Upon further research and application, it became apparent that the data dimensionality reduction is sensitive to the parameter settings and must be fine-tuned carefully to be successful.