i am trying to predict laptop prices from previous sales. based on a text description of the laptop condition and the sale price. the affect of the condition on the final price is going to be different for a high value laptop and low value laptop. e.g. a laptop with a broken screen may be worth 25% of a laptop with in excellent condtion. can i feature scale each laptop model based on the range of sale value for the model and then try and predict the scaled value from the text description in my test set. and then transform back to get the predicted price. or is there a better way to capture the fact that the affect of a condition variable will be different for depending on some baseline value for the model.
Asked
Active
Viewed 14 times
