Website (SPA) for visualizing the distribution of empirical eigenvalues of covariance matrices

Page: https://krzemon.github.io/eigenflow/

The website allows you to generate an estimator of empirical covariance matrices with different dimensions and properties in the form of a probability density distribution of eigenvalues, together with a theoretical curve describing this distribution and providing statistics describing the distributions.

Applications of Covariance Matrix Spectrum Analysis

The analysis of the spectrum of covariance matrix estimators has many practical applications in fields such as:

• machine learning
• physics
• biology
• finance

In general, it is used in areas involving multivariate data analysis.
In simple terms, the obtained spectrum helps determine which components of the matrix represent meaningful information and which correspond to noise.

Frontend

The frontend is implemented as a Single Page Application (SPA) built with:

• HTML
• CSS
• JavaScript

Charts and visualizations are generated using Chart.js (chart.umd.min.js), including histograms and theoretical distribution curves.

The application retrieves data from a backend API, which provides the required data in JSON format.

Backend repository: https://github.com/Krzemon/random-matrix-api

Static frontend files are hosted on GitHub Pages, allowing the application to be publicly accessible in a web browser without requiring a dedicated server.