R Linguistic statistical analysis technology research on the key points of eigenvalue selection

Key points of eigenvalue selection technique

Wang Limin

Article Source: Network

1. Characteristic values

Eigenvalue is an important concept in linear algebra. In mathematics, physics, chemistry, computer and other fields have a wide range of applications. Set A is an n-order matrix, if there are several m and not 0 N Willi Vector x, so that AX=MX is established, it is said that M is a eigenvalue or intrinsic value. A non-0 n Willi Vector x is called a eigenvector or eigen vector of the eigenvalues m of the matrix A, referred to as a eigenvector or a eigen vector.

2. Feature selection using the R language Boruta package

In the process of data analysis, the use of various charts for data exploration is necessary for the pre-work. Descriptive statistics include histograms, scatter charts and other tools to explore continuous data, you can use bar chart, cross-grouping table and other tools. The so-called "pivot table" in Excel is actually an interactive cross-grouping table.

3.R language extraction of time series eigenvalue

There is a set of one-minute series of data (a variable), think about extracting the characteristic parameters of this time series, and as the input of the clustering algorithm, realize the classification of this set of data.

4. Finding eigenvalues and eigenvectors of matrices

The function Eigen (SM) is used to calculate the eigenvalues and eigenvectors of the matrix SM. The return value of this function is a

Contains a list of values and vectors two components. Command

> EV <-eigen (Sm)

This list will be assigned to EV. Ev$val represents the eigenvalue vector of SM, and Ev$vec is the corresponding eigenvectors.

One of the matrices. Assuming we only need eigenvalues, we can use the following command:

> Evals <-eigen (Sm) $values

Evals now has eigenvectors, while the second component is discarded. If you use the following expression as a

Command

> Eigen (Sm)

These two components, together with their names, will be displayed. For large matrices, if not necessary, it is best not to use the following

Expressions to evaluate eigenvectors

> Evals <-eigen (Sm, only.values = TRUE) $values

R Linguistic statistical analysis technology research on the key points of eigenvalue selection