I was working on a similar system for more general reviews and found that compared to fraudulent positive reviews negative reviews have a lot less incentives ex

I was working on a similar system for more general reviews and found that compared to fraudulent positive reviews negative reviews have a lot less incentives except against public figures and tend to be more honest as a result. With that said, protecting both the submitter and the accused of defamation can be accomplished partially with community moderation and aggressive redaction of as many identifying names as possible - companies are, however, not exempt (in fact, the primary key that I used for complaints). And to avoid unfair hive mind viral hits on characters, any review beyond a certain community reputation needs to be reviewed for validity and made clear to the community that the story is an allegation. Furthermore, I wanted to encourage more of a dialogue and similar to Yelp reviews, so if you are given a negative reputation you must be provided a means to defend yourself - and the status of defense must be made clear as well. For example, a statement can be made public, private (only to the person that originally wrote the review), “no comment,” or no response. If someone didn’t originally provide their contact info out of fears, the community should be made clear that it’s anonymous and will always remain so. However, protecting the information at your discretion until a certain reputation has been reached should provide more credibility as well. The hard part was trying to figure out how to contact the accused at a corporation for low-touch interactions because we tend to not have the same person repeatedly (hence why normally I just used the company, not individuals).