Aster: Principal Component Analysis And Unsupervised Machine Learning

[texas]

Background:  A process used to emphasize variability and bring out strong patterns in a dataset.  This variability is expressed by principal components; which are directions of highest degree of variance.  The first several principal components represent 80-90% of the variance and hence most important.

Use Cases:  

  -  Dimensional Reduction / Compression / Image Recognition

 -  Medical Diagnosis / Medical Imaging / Sensor Data

  -  Outlier Detection

[texas]

https://www.youtube.com/watch?v=FJcPLXVFZeU