How to convert a set of probabilities to a value between 0 and 1?

450 Views Asked by At

I need to make a function which output a value between 0 and 1 which specifies the quality of the quantity I am measuring (0 is bad quality and 1 is high quality). The input to this function are 3 probability values which vary from 0 and 1. The first probability value says the goodness of my measurement and second and third probability values are the correct measurement rate and wrong measurement rate. I need to use this information (probabilities) to make a model/function that gives the current quality of my model. So how to normalise these 3 probabilities to a value between 0 and 1. Any hints would be really helpful. Thanks.

1

There are 1 best solutions below

0
On

One way to do it is using generalized (power) mean. It is an appropriate solution if your $3$ values are (almost) equally significant. Be aware that for negative power this mean has problem with zeros.

Another way is using weighted mean. In this case you can give weights according to values significance.

Also it is possible to combine both ideas.