Although the calculation of trust score is
clearly mentioned by theymos, it is still complicated for some users. I was also confused till yesterday. After spending some times, I was able to understand how trust score is calculated.
Here I will try to present the calculation in an easy way so that everyone can get it easily.
Before going to the calculation, let's understand what are elements shown in trust. Have a look at the below image and it will be clear for you.
In trust score, there are 3 score-
1. Positive Scores
2. Negative Score
3. Undefined Score
Skipped the neutral.
Now I will show how those scores are calculated by using some users data.
This calculation is applied when a user never received a negative feedback from DT.1. When a user receives a positive feedback, he gets +1 in total positive feedback but the score remains 0. It starts changing after approximately 2 weeks, theymos uses "round_up" for months (Special thanks to LoyceV). User gets 1 point after each round_up month upto 10.
Since I already have calculated
Vod's trust score, I will use Vod as en example here.
Total Unique Positive Feedback- 15
Total Unique Negative Feedback- 0
Out of his 15 positive feedbacks, he receives 13 feedbacks which have a weight of 10 months.
a. The output= 13*10=130
b. Received positive feedback from MadZ 8 months ago, output=8
c. Received positive feedback from ibminer 4 months ago, output= 4
Total positive score(a+b+c)= 130+8+4= 142
A positive score can also be gained if the user receives negative feedback. However, I will show here all the other negative and undefined too since those are calculated and related one with another algo.
The following is applied when user receive both negative and positive feedback-
score = unique_positive - 2^(unique_negative)
if score >= 0
start_time = time of first negative
score = unique_positive since start_time - unique_negative since start_time
if(score < 0)
return
(orange)
Information needed-
a. Total Unique Positive Feedback
b. Total Unique Negative Feedback
c. total Unique Positive Feedback after the first negative feedback
User- koshgel
Score= -2a. Total Unique Positive Feedback= 6
b. Total Unique Negative Feedback= 3
c. total Unique Positive Feedback after the first negative feedback= 0
Score= a - 2^b (2 to the power unique negative)
=6 - 2^3
=6 - 8= -2
Now, if the score was greater than or is equal to 0; score>=0, we would apply the following method which has been applied for user Anduck
User- Anduck
Score= 1a. Total Unique Positive Feedback= 9
b. Total Unique Negative Feedback= 1
c. total Unique Positive Feedback after the first negative feedback= 2
Score= a - 2^b (2 to the power unique negative)
=2 - 2^1
=2 - 2=0
Since the score is
greater than or is equal to 0,
Apply
score = unique_positive since start_time - unique_negative since start_time
where start_time= First Negative Feedback
Score= c - b
=2 - 1= 1
Now, if the output would smaller than 0 or negative, the trust score would be undefined. Here is the trust score calculation of user zeeshanblcUser- zeeshanblc
Score= a. Total Unique Positive Feedback= 2
b. Total Unique Negative Feedback= 1
c. total Unique Positive Feedback after the first negative feedback= 0
Score= a - 2^b (2 to the power unique negative)
= 2 - 1 = 1
Since the score is greater than
or is equal to 0,
Apply
score = unique_positive since start_time - unique_negative since start_time
where start_time= First Negative Feedback
Score= c - b
= 0 - 1
= -1
Since the score is smaller than 0, the trust score is undefined.
Color of Trust Scores1. Negative Score- Red
2. Undefined Score- Orange
3. Positive Score-
a. Initial Score- Black
b. When trust score reaches at 5, it turns into Light Green and remains same till 14.
c. When trust score reaches at 15, it turns into Dark Green
Special Thanks to-asche for fixing the errors.
LoyceV for explaining the color of trust scores.
