Automattic’s Akismet has been long known as the king of spam detection. It has been ported to multiple content management systems like WordPress, Drupal, MovableType, phpBB, Joomla, b2evolution, Geeklog, and many more. It is, hands down, one of the most popular solutions to handle all this spam, but now there is something else ready to take on the likes of Akismet. It, perhaps, might even be more effective than Akismet.
Defensio, a comment spam detection system which claims 99.77% accuracy, is really striving to become serious competition to Akismet. Not only that, but the development team aims to pick up where Akismet has left off.
The system improves upon what Akismet has built in many ways: the system ranks comments by “spaminess,” it adapts to your blog over time, and it also provides RSS feeds for your comments. But it also does so much more.
Rick Olson of ActiveReload discussed the ways in which Defensio differs from Akismet:
How does Defensio differ? It sports a very similar API with a few added methods. First, it can scan articles as they’re posted. I’m assuming this provides an extra layer of protection by finding topics that may be off topic. They also provide a method to retrieve basic statistics. Though I think the most important additions are the inclusion of unique signatures and ‘spaminess’ scores. Unique signatures allow me to report spam/ham on previously checked comments without having to reconstruct the original request again. The spaminess value provides a convenient way to sort comments. Perhaps you want to see valid comments with the highest spaminess value, or spam with the lowest spaminess value to look for false reporting.
With Defensio’s claims of 99.77% accuracy, I managed to get around 96.43% accuracy (I took this screen-shot this morning, but it has changed to 96.43% since then) , but I did not reach a level of comments which was sufficient to accurately judge. If I had given it a few more days, I quite possibly would have over 99% accuracy, but I just managed to get hit with a false-positive yesterday. I believe it is still in “training.” Over the past four weeks, I received zero spam on my blog and two false-positives. With that in mind, I believe it was fairly accurate considering the low volume of comments. The aggressiveness of the filtering can also be tweaked. I set mine to a relatively low setting, but you might prefer to set it higher once you have trained the application.
The only suggestion I have would be to send email notifications when there are questionable items in the spam queue. But other than that, I was more than satisfied with the product, and I would recommend that people who have not been overly satisfied with Akismet to give it a shot. It is a really simple setup.
If you have used Defensio or have any questions, please let me know in the comments section. I’ll be glad to hear about your experiences and offer any help that I can.