This is how we tell if a post is created from an automated content spammer.
Before any post is created, we check that the post does not match any pattern associated with an automated content spammer. Such patterns include a large amount of links in a single post. If a pattern is matched, the post is flagged for checking. We also select a number of random posts that do not fit any patterns, and check them too.
With the flagged posts, we check from various spam databases across the web to see if the username/email/IP has ever been used in the past to spam a forum. If a history is found of spamming, that IP address will get banned from our websites.
Our Administrators routinely monitor posts which get flagged. This ensures that all bans are legitimate.
If the system picks up an innocent user, we have a full appeal system which the user can apply for. This system is regulated by our Administrators.