Analyzing and studying Dimensionality Reduction Techniques for High-Dimensional Data
Author(s):
Rutuja P. Deshmukh, Shweta A. Jagdale, Satish S. Banait, Akash R. Kadlag, Pranjali A. Baviskar
Keywords:
Big Data, Dimensionality Reduction, High-dimensional data, ICA, Mean-Squared Error, NMF, PCA, SVD, Truncated-SVD
Abstract
In the fast-moving world, data is accumulating at an unprecedented speed from vivid sectors across the sphere such as micro-array gene expression data, medical data, ECG and MEG data research, satellite images, IoT devices, etc. is considered as high dimensional data. This data has a lot of features and thus directly affects the output of machine learning algorithms at an exponential rate. Thus, dimensionality reduction (DR) helps to solve the problem of the curse of dimensionality by extracting the relevant features without forfeiting the useful data. The purpose of this research is to compare and analyze different dimensionality reduction techniques namely Principal Component Analysis (PCA), Independent Component Analysis (ICA), Singular Value Decomposition (SVD), Truncated-SVD and Non-negative Matrix Factorization (NMF) on Imagenet dataset (unsupervised dataset) for five different values of components - 40, 45, 50, 55 and 60 each. These algorithms are examined on the basis of execution time, accuracy of dimensionality reduction techniques and load analysis, that is, Mean Squared Error (MSE). The algorithm with the least execution time and number of components giving the most information is concluded as a suitable algorithm for dimensionally reducing high-dimensional data.
Article Details
Unique Paper ID: 154651

Publication Volume & Issue: Volume 8, Issue 11

Page(s): 828 - 833
Article Preview & Download


Share This Article

Join our RMS

Conference Alert

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024

Last Date: 15th March 2024

Call For Paper

Volume 10 Issue 10

Last Date for paper submitting for March Issue is 25 June 2024

About Us

IJIRT.org enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on editor@ijirt.org

Social Media

Google Verified Reviews