I am trying to convert this adaptive Bayesian rating formula to PHP code: see here.
The following are details of the various parts of the formula.
- nvotes: total votes so far
- nlinks: total number of links
- nvotes (k): the number of votes cast for the r-th channel.
- deltarank (k, m): rank increment caused by the k-th vote, which is called on the m-th channel.
- nsaves (i): the number of users who keep the link to the link to their linkibol.
- a: save exponent (ad-hoc value close to 1)
- age (i): difference (in days) between the added date link and the current date.
- b: decay rate (ad-hoc value close to 0)
(full information about the formula can be found at http://blog.linkibol.com/2010/05/07/how-to-build-a-popularity-algorithm-you-can-be-proud-of/ - scroll down to section "How do we introduce popularity in linkibol?" section)
I can easily convert most of this function to PHP code, but a bit that I don't understand is sigma and deltarank. I am not sure what this bit should do or what values to pass k and m.
If anyone has any advice or can break the difficult bit of the formula, which would be great, then I can see what would be the best way to implement it in PHP - there might be functions that I could use, etc. ..
Richw source
share