Use values of time vector as features or use PCA

Vote count:

0

I have a vector of the form

t=[value1,value2,...,valueN]

where t goes from 1 to 100. This is a 2 dimensional vector, it can be plotted against an axis X. I want to match every value of t (1-100) to a single class, for machine learning classification.

My question is what is the optimal course of action in this case, There are 2 options

1) use PCA for a transformation from 2D to 1D
2) use the values of each time value as features for the class.

Can someone please explain the advantages and disadvantages on each case?

Thank you.

asked 1 min ago

user3025898

12

Use values of time vector as features or use PCA