How does finding duplicate articles work?
I would suggest to compare (also) URL where the articles point. I am subcribed to some feeds which also my friend uses on Google. And I am subscribed to his "shared feed", so some articles are also from there.
Also I have some newspapers feeds, which have everyday similar titles for some articles, but every day it is different story and has different URL, but FeedReader keeps showing them as duplicates...