You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The current predictions are pessimistic because they reflect the expected ratio of open to closed sections. They should instead be based on the expected probability that there are any open sections of each type.
Implementation-wise, the most natural way to do it would be to just take the ceiling of the ratio and run it through the same calculations as it does now. But since there are only a few semesters recorded so far, it seems like the results would be fairly rough and inaccurate. Maybe averaging over a period of a week would continue to help smooth it out.
This wouldn't be perfect because it doesn't reflect that the only available section might conflict with the user's existing schedule. But it's probably better than saying that 50% closed = 50% chance of getting in. Users are more likely to be searching for classes they really need, in which case they would likely rearrange their schedule around the only available section if necessary.
The text was updated successfully, but these errors were encountered:
The current predictions are pessimistic because they reflect the expected ratio of open to closed sections. They should instead be based on the expected probability that there are any open sections of each type.
Implementation-wise, the most natural way to do it would be to just take the ceiling of the ratio and run it through the same calculations as it does now. But since there are only a few semesters recorded so far, it seems like the results would be fairly rough and inaccurate. Maybe averaging over a period of a week would continue to help smooth it out.
This wouldn't be perfect because it doesn't reflect that the only available section might conflict with the user's existing schedule. But it's probably better than saying that 50% closed = 50% chance of getting in. Users are more likely to be searching for classes they really need, in which case they would likely rearrange their schedule around the only available section if necessary.
The text was updated successfully, but these errors were encountered: