Floorball is a type of floor hockey. During the game, substitutions can be made.

The team is also allowed to change players any time in the game; usually, they change the whole team. Individual substitution happens sometimes, but it usually happens when a player is exhausted or is hurt.

I would like to use an RNN to predict when the next substitution will happen for a team. However, I have no pre-existing dataset to train on. Is there a way that I can start predicting without a dataset and continually improve accuracy as more games are played?

  • 1
    $\begingroup$ The two questions you ask in the title seem to be independent questions. They should probably be asked in separate posts. In the post about floorball, you should give or link to a description of the game. Welcome to AI.se! $\endgroup$ Mar 17 '19 at 18:09
  • $\begingroup$ @PhilipRaeisghasem Thanks for the feedback and welcoming me! I tried to narrow it down and removing multiple questions. $\endgroup$
    – J.norberg
    Mar 17 '19 at 20:30
  • $\begingroup$ I think you might have some misconceptions about what training entails. "A dataset" would be a set of inputs that a neural network could use to optimize its weights in order to do well at a certain task. This could be information about many past games and when substitutions happened in them. Do you have such a dataset? Or are you asking if you can do prediction while gathering the data? $\endgroup$ Mar 17 '19 at 20:51
  • $\begingroup$ @PhilipRaeisghasem The first match being played there exist no dataset. I want it to continually update the dataset every time a substitute is made. After the dataset is updated I also want it to predict the next substitute. When the match is finished the dataset is saved and used in the next match. Then it grows every match and gets more and more accurate. Thank you very much in advance! TL DR "Do you have such a dataset?" Not at the beginning! "Or are you asking if you can do prediction while gathering the data?" yes! :) $\endgroup$
    – J.norberg
    Mar 17 '19 at 22:32

Your Answer

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

Browse other questions tagged or ask your own question.