Case Study
How Areto’s context-detecting software compares to social media platforms’ banned words lists
Social media platforms allow users to upload banned words lists to automatically block comments.
Are they accidentally blocking the sweet stuff along with the sour?
We took the profanity test to find out!
SNAPSHOT:
Analysis: We put a 1,000-word “banned words” list head-to-head against Areto’s AI systems
Data: Over a 30-day period, out of 95,000 comments we collected and analyzed from X, Instagram and Facebook, we observed 4,710 instances of profanity.
RESULTS:
BANNED WORDS
4,710 profanity found
4,710 profanity actioned
ARETO
4,710 profanity found
1,509 profanity actioned
3,201
false positives found by Areto
68%
positive comments not blocked by Areto
+445
Areto ID’d additional abusive comments that didn’t contain profanity for a total of 1,954 actioned comments
SUMMARY:
Areto outperforms banned words lists, finding more abuse and saving genuine fan engagement
Thanks to Areto's AI systems, Areto outperformed the banned words list, saving 3,000+ fan comments from the chopping block, & finding 445 more abusive comments that didn't include profanity.
CONCLUSION:
CONTEXT IS QUEEN when it comes to effectively moderating online abuse & harassment and promoting healthy dialogue, fostering authentic engagement and nurturing passionate fandom.