I have a very basic algo wherein each person gives an opinion about an article which has a score. All opinions add up points to the article score. For example, If person says I like it, article gets 5 points and likewise, -5 for dislike. In addition to this, there are some additional points based on person's rating. For example, if a person has good rating himself, we assign 25 additional points to that article. so if that person likes the article, we give 30 points instead of 5.
in the end I want to calculate score between the range of 1%-100%..but as you can see, we can't use simple percentage formula here...let's say if we have 5 people with good ratings and they all like that article, score will become 150 and if we divide that with 5 answer will be 50 which doesn't make much sense. Is there any way or formula for such variable scores? If it was as simple as 5 points for each like then I could have used (5/25)*100 = total %age but it won't work in my case.
Honestly speaking, I don't know where to start this so I haven't done anything as of yet. Sorry for my dumbness.
The percentage is a function of the ratio between actual score and possible score based on involved voters so far.
The full range is then calculated as:
The actual value you already know, so now you need to decide if you scale the range from 0% to 100%, or -100% to 100%, and thus how much to scale the actual value by to get your final percentage.