Scribd
Upload
9 views1 page

PCA Notes

Principal Component Analysis (PCA) is a dimensionality reduction technique that transforms a dataset into principal components by projecting data onto new orthogonal axes that capture maximum variance. The process involves normalizing data, computing orthonormal vectors, sorting components by variance, and selecting the top components while discarding low-variance ones. While PCA reduces dimensionality and improves performance, it can be hard to interpret and may lead to information loss.

Uploaded by

Muhammed RAYAN
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views1 page

PCA Notes

Principal Component Analysis (PCA) is a dimensionality reduction technique that transforms a dataset into principal components by projecting data onto new orthogonal axes that capture maximum variance. The process involves normalizing data, computing orthonormal vectors, sorting components by variance, and selecting the top components while discarding low-variance ones. While PCA reduces dimensionality and improves performance, it can be hard to interpret and may lead to information loss.

Uploaded by

Muhammed RAYAN
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Principal Component Analysis (PCA / K-L Method) PCA is a dimensionality reduction technique that

transforms a dataset with n attributes into k new variables (k ≤ n) called principal components. Key
Idea: PCA projects data onto new orthogonal axes that capture maximum variance. Steps: 1.
Normalize data so all attributes are in the same range. 2. Compute orthonormal vectors (principal
components). 3. Sort components based on variance (importance). 4. Select top k components and
discard low-variance ones. Important Points: - Components are orthogonal and independent. - PCA
combines attributes instead of removing them. - Helps reveal hidden patterns and structures.
Advantages: - Reduces dimensionality - Improves performance - Removes redundancy
Disadvantages: - Hard to interpret - Possible information loss Conclusion: PCA reduces data size
while preserving important information by projecting data onto directions of maximum variance.

You might also like