Sentences found in the text being translated that are retrieved by the Translation Memory because a similar content has already been translated in the past. They are usually divided into percentage groups (50%-99%) and collectivelly called Fuzzy Matches.