DIMENSIONALITY REDUCTION BY USING EIGEN VALUE AND EIGEN VECTOR

Abstract

The main objective of this study is to convert a high dimensional data into a low dimensional data by preserving the important information. In modern dataset, data contain a hundred or thousands of variable which make difficulty in computing and the interpretation becomes more complicated. Firstly, give a basic summary of the difficulties that we face in case of high dimensional data, like redundant or irrelevant information than can reduce the performance of machine learning model and statistical models. In this report we learn that dimensionality reduction techniques can helps to overcome these problems by transforming the original Data into low dimensional data by using mathematical tools that are eigen vectors and Eigen values, which forms the basis of techniques such as Principal Component Analysis (PCA). Then we discussed the literature review of dimensionality reduction using Eigen value and Eigen vector.