fwiw i found that a large proportion of the metacritic user reviews i looked at the night it come out appeared to be fake, not all containing racist/sexist cont

fwiw i found that a large proportion of the metacritic user reviews i looked at the night it come out appeared to be fake, not all containing racist/sexist content, and more seemed to take a "tolkein would be rolling in his grave" slant it was a fun problem to consider though, how to evaluate whether a review is fake or not. obviously the most powerful tools would be ip logging or browser fingerprinting or requiring other user verification but absent that i wonder what approaches review verification services take based on the content alone. i use some plugin for amazon reviews that i think looks for similar terms but it seems like it can sometimes be overzealous when products are very simple so many reviews seem similar when they might be from unique people