Iterative Aggregation Method for Solving Principal Component Analysis Problems

TL;DR

This paper introduces a two-level aggregation method to efficiently solve PCA eigenvalue problems, especially for large text datasets, by leveraging a multilevel approach inspired by structural analysis techniques.

Contribution

It presents a novel two-level aggregation approach for PCA that improves computational efficiency in large-scale data analysis.

Findings

Effective on large text datasets

Reduces computational time

Maintains accuracy of PCA results

Abstract

Motivated by the previously developed multilevel aggregation method for solving structural analysis problems a novel two-level aggregation approach for efficient iterative solution of Principal Component Analysis (PCA) problems is proposed. The course aggregation model of the original covariance matrix is used in the iterative solution of the eigenvalue problem by a power iterations method. The method is tested on several data sets consisting of large number of text documents.