How about a way to report, "I don't like receiving this type of message" and "I like this type of message", along with an ML system that can learn which type of

How about a way to report, "I don't like receiving this type of message" and "I like this type of message", along with an ML system that can learn which type of users send lewd messages, and which people either like or dislike receiving them? The men who send DTF messages eventually only get paired with women who are interested to receive them. With smart ML, the site is what each person wants it to be. Beyond categorizing users, perhaps you can categorize the sentiment of messages themselves using unsupervised learning on the messages and user responses. Or maybe just a star rating for messages, Uber style. Drive everyone who wants to send DTF messages to another site.