Pca and eigenvalue interpretation

In summary: So you might have to use both.In summary, the conversation discusses the use of PCA to analyze the movement of human walking. The speaker outlines the steps they took, such as dividing the body into parts, filming and tracking the x position of each part, and creating a matrix to find the PCA. The question is then asked about how to interpret the data and the meaning of eigenvalues and the correlation between the first and second principle components. The speaker also asks about other possible analyses that can be done with PCA. Finally, they mention the use of kinematics in conjunction with PCA for modeling joints and nodes of the body.
  • #1
martinbandung
7
0
hello, i have a reasearch to analyse the movement of human walking using pca. i did it like this
1. i dibide the body into some part (thigh, foot, hand, etc)
2. i film it so i can track the x position of the parts
3. i get the x to t graph for every part
4. i make a matrix which column is the position x for every parts, and rows is the position for each time
5. i gind the pca of the matrixquestion
how do i interpret the data?
what is the meaning of the eigen values?
what correlation berween first principle component and second principle component?

what further analysis can i get from doing pca?

thanks for the answers
 
Physics news on Phys.org
  • #2
Hey martinbandung.

I think you should outline the PCA technique you are using.

If you are modeling joints and "nodes" of a body then you use something called kinematics. Forward kinematics is when you adjust each node for a new position based on how all "nodes" move and inverse kinematics is when you solve for the "node" positions given final positions for some of the "nodes" (and then solve for the remaining ones).

The technique is covered in a lot of computer animation and physics books.

Principal Component Analysis (PCA) is a statistical technique and is separate from kinematics which is a physics topic.
 

1. What is PCA and how does it work?

PCA (Principal Component Analysis) is a commonly used statistical technique for dimensionality reduction in data analysis. It works by finding the most important features, or principal components, in a dataset and transforming the data onto a new coordinate system based on these components. This allows for a simplified representation of the data, making it easier to analyze and visualize.

2. What is the purpose of eigenvalues in PCA?

Eigenvalues represent the amount of variance explained by each principal component in PCA. They are used to determine the relative importance of each component in the data and aid in selecting the appropriate number of components to retain for analysis.

3. How do you interpret eigenvalues in PCA?

The larger the eigenvalue, the more important the corresponding principal component is in explaining the variation in the data. Eigenvalues can also be used to calculate the proportion of variance explained by each principal component, with a higher proportion indicating a more significant component in the dataset.

4. Can eigenvalues be negative in PCA?

No, eigenvalues in PCA are always positive. This is because they represent the amount of variance explained by each principal component, and variance cannot be negative.

5. How do you determine the number of principal components to retain in PCA?

There are a few methods for selecting the number of components to retain in PCA, such as the Kaiser rule, scree plot, and cumulative variance explained. These methods use the eigenvalues and/or proportion of variance explained to determine the optimal number of components for analysis.

Similar threads

  • Set Theory, Logic, Probability, Statistics
Replies
29
Views
6K
  • Set Theory, Logic, Probability, Statistics
Replies
1
Views
2K
  • Set Theory, Logic, Probability, Statistics
Replies
4
Views
2K
  • Linear and Abstract Algebra
Replies
4
Views
2K
  • Set Theory, Logic, Probability, Statistics
Replies
3
Views
3K
  • Engineering and Comp Sci Homework Help
Replies
5
Views
2K
  • Linear and Abstract Algebra
Replies
2
Views
869
  • General Discussion
6
Replies
190
Views
9K
  • Advanced Physics Homework Help
Replies
5
Views
1K
  • Linear and Abstract Algebra
Replies
5
Views
1K
Back
Top