I forgot to link in the OP. Then remembered, and forgot again.
Something of interest: Jeffery’s interval. Using the lower bound of a credible interval based on that distribution (which is the same as yours) will probably give better results than just using the mean: it handles small sample sizes more gracefully. (I think, but I’m certainly willing to be corrected.)
This seems to use specific parameters for the beta distribution. In the model I describe, the parameters are tailored per domain. This is actually an important distinction.
I think using the lower bound of an interval makes every item “guilty until proven innocent”—with no data we assume the item is of low quality. In my method we give the mean quality of all items (and it is important we calibrate the parameters for the domain). Which is better is debatable.
I forgot to link in the OP. Then remembered, and forgot again.
This seems to use specific parameters for the beta distribution. In the model I describe, the parameters are tailored per domain. This is actually an important distinction.
I think using the lower bound of an interval makes every item “guilty until proven innocent”—with no data we assume the item is of low quality. In my method we give the mean quality of all items (and it is important we calibrate the parameters for the domain). Which is better is debatable.