0
$\begingroup$

Say I have 1000 graphs that shows sales every year for the last 10 years for 1000 different companies. And say each of those graphs belong to either domestic countries or foreign countries.

Is it possible I could input the different graphs into a classifier? That is, could a model predict based on the graphs whether or not the country was domestic or foreign? If so, how would you do that in python or r?

$\endgroup$
0
$\begingroup$

Yes, the graph coordinates can be used as features for training your algorithm( whichever classification algorithm you choose). Features will be X and Y and output will be labelencoded type of country.

$\endgroup$
1
  • $\begingroup$ That's kind of what I figured but I'm having trouble visualizing what that would look like. Is representing each graph literally just like a vector of x coordinates, a vector of corresponding y coordinates, and then a vector of corresponding factors (i.e. 1 = domestic, 0 = foreign) for each of those coordinates? And then just repeat for each different graph, then split it up into the trainset and testset? Wouldn't the predictions also just be whether or not each individual coordinate is a 0 or 1? Like, is there a way to predict whether an array of values would be classified as a 0 or 1? $\endgroup$
    – BigNate
    Oct 5 '19 at 22:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.