You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
"Roommates" allows its users to select their hobbies/interests. The selection is like "yes/no" question and the matching algorithm that calculates the match score of two people is based on the number of in common attributes.
However, I am not sure that this model reflects real life since there might be blockers e.g. vegan and carnivore.
In my opinion, we should change the "yes-no" questions to "I am - don't care - no way (blocker)".
And the matching algorithm will be modified as well according to the following matrix-
Person A/Person B
I am
Don't care
No way
I am
+1 to the matching sum
+0 to the matching sum
Matching result is set to 0
Don't care
+0 to the matching sum
+0 to the matching sum
+0 to the matching sum
No way
Matching result is set to 0
+0 to the matching sum
+1 to the matching sum
The text was updated successfully, but these errors were encountered:
Hi,
Not sure about the score, but I think it's a good idea. In any case, it should be that the scoring-function is configurable (injected to the matching-function).
"Roommates" allows its users to select their hobbies/interests. The selection is like "yes/no" question and the matching algorithm that calculates the match score of two people is based on the number of in common attributes.
However, I am not sure that this model reflects real life since there might be blockers e.g. vegan and carnivore.
In my opinion, we should change the "yes-no" questions to "I am - don't care - no way (blocker)".
And the matching algorithm will be modified as well according to the following matrix-
The text was updated successfully, but these errors were encountered: