I am dealing with a problem where the output of my model, can have numbers like 1-3000 (around) (score in a game). This is like a score in a game. Giving a least squared error setting, for a model, where I want to try to predict the score, will make me end up with a large error, and will not be quite reflective of a good prediction. For example, for a game of score 2800, I might be happy with a prediction of 2100, but for a game with score 300, I definitely will not be happy with a prediction of 1000, even though both have the same offset of 700.

At the same time, if I have a classification like prediction over all the 3000 classes (let's define each discrete score in the range to be a class) and then give my results as a discrete probability distribution over the scores, then I am not using the closeness of scores in the picture, because the model would assume all classes are independent, but a score of 1900 is more close and similar to a class of score 2000 than a class of score 1.

So, how exactly should I come up with a method to learn and evaluate , with a good interpretable ground truth when I am trying to build a model to predict such data (like game scores)?

Thanks in advance for all help.

  • 2
    $\begingroup$ I would formulate this as a regression problem. $\endgroup$ Jun 18 '15 at 6:59
  • $\begingroup$ What is your features? $\endgroup$ Jun 18 '15 at 19:53

What you suggest implies to estimate a gigantic number of parameters to say the least. Unless you have a more than huge dataset you should avoid this technique.

Why not simply log-transform your score ? It follows that you consider the size of the ratio on your score. 2100 for 2800 would be as bad as $700*21/28=525$ for $700$ which seems more appropriate.

If you have scores with 0, I would (unconfidently) suggest to translate the whole scoring from 1 to avoid a problem.


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.