1

I am using perceptron machine learning to solve the binary classification problem A vs B. For this I have to assign the actual values of A and B to either 1 or -1 to be able to use perceptron. Does it matter which value I assign, as it is not clear which value makes more sense to assign 1.

user100000
  • 11
  • no it does not matter – Alberto Mar 20 '23 at 20:07
  • 1
    Instead of "A" and "B," why not call the classes "Abel" and "Baker"? Or how about renaming them for the second letters of those names, "B" and "A," respectively? When you do that, what happens to the $\pm 1$ values? – whuber Mar 20 '23 at 20:28

1 Answers1

0

There is a "yes" and a "no" to this.

NO

These are just ways of encoding categorical information as numbers that math can handle. In other implementations, you will encode as $0$ or $1$, as is discussed in some nice answers to a question of mine from last month.

YES

There will be an output from your model, and you need to know how to interpret that output. For instance, if your output is the probability of membership in the positive class, it matters which category is the positive class.

Dave
  • 62,186