Please click on a link below to get more details.
Trackback is one of three types of Linkbacks, methods for Web authors to request notification when somebody links to one of their documents. This enables authors to keep track of who is linking to, or referring to their articles.
Spammers are extensively abusing the TrackBack feature to insert spam links in blogs (sping). Trackback spam was widely adopted by spammers because it was able to avoid some of the safeguards designed to stop normal comment spam (using the comment form). However not any longer.
Trackback spams are on the rise and comprise the second most source of comment spams in Wordpress blogs.
Comment Guard Pro employs an array of technologies to directly identify and eliminate Trackback comment spam, coded in two pluglets - Trackback Link Validator and Trackback Content Validator. In addition to regular validation of reciprocal link, Trackback Link Validator pluglet also validates the authenticity of the link. Trackback Content Validator checks various aspects of contents to eliminate spam trackbacks. Both of them have been extensively tested with live data over several months to prevent any occurrence of false positives. Any comment which passes these two pluglets is then checked by other pluglets too, thereby providing very strong protection against Trackback spam. Additionally we re-check trackbacks with a configurable time-delay to eliminate attempts by spammers to trick using temporary linkbacks etc. In several months of tests not one spam trackback slipped through our trackback spam protection while not a single genuine trackback was stopped.
Referer spam is a kind of spamdexing (spamming aimed at search engines). The technique involves making repeated web site requests using a fake referer url that points to the site the spammer wishes to advertise. Sites that publicize their access logs, including referer statistics, will then end up linking to the spammer's site, which will in turn be indexed by the search engines as they crawl the access logs.
Referrer Blacklist pluglet provides protection against comment spam which comes with a referrer payload by allowing you to easily blacklist keywords in referrer url. For example you can blacklist comments with referrer pointing to url's containing words like casino, gambling etc. Referrer Blacklist pluglet ships with a list of common keywords, which you can add/edit/delete to your liking.
Comment Guard Pro was created because all existing plugins we tested were either ineffective because of high false negatives or because of very high false positive rates. Checking the spam queue to flush out false positives is ineffective when you have very large number of comments in queue. In a popular anti-comment spam plugin we tested we literally had tens of thousands of spam in queue, imagine shifting through them for valid comments! On the other hand false negatives allow spam comments to get approved for publication on your blog, often inappropriate for general audience.
We extensively tested to fine-tune the default configuration to minimize false positives (where a comment is falsely identified as spam) as well as false negatives (where a spam is misclassified as genuine comment).
How effective are we?
In 375 days 23 hours and 29 minutes of testing (at the time of writing) we had zero false positives and zero false negatives in Simple Thought blog. We had similar success with our other blogs too.
An expert system is a software system that incorporates knowledge & concepts derived from experts in a field and uses that to provide solution to users of the software.
Comment Guard Pro incorporates several years of experience in comment spam detection, both ours and the internet community at large. Comment Spam is a moving target, so the solution also needs to be adaptable.
Expert System, also known as a knowledge based system, allows us to incorporate various good technologies for comment spam detection, each of which tackles either a specific sub-class of comment spam or checks for certain defining characteristics within a larger class. In combination they form a formidable force against all kind of comment spams. For example, Are You Human pluglet provides full protection against robotic comment spams using the comment form. However it doesn't address trackback spams, we have Trackback Link Validator pluglet as well as Trackback Content Validator pluglet to provide strong protection against trackback spam. On the other hand, Blacklist Keywords pluglet checks all types of comments looking for specific keywords which clearly indicates comment spam.
Incorporating an Expert System for comment spam detection provides us with immense versatility and adaptability in fighting a moving target (comment spam), safe detection (very low false positives and very low false negatives) of which has eluded bloggers so far. In extensive tests with Comment Guard Pro spanning over a year and processing over a million comments, we had very good results (zero false positives and zero false negatives) so far.
The Expert System also provides unprecedented flexibility & configurability in formulating your own strategy against comment spam, should you wish to do so. You decide the weight (value) and decision making ability of each rule / module. You can also incorporate your own spam detection rules within Comment Guard Pro in the form of pluglets. In short you are in full control over your comment spam fighting initiative.
Traditionally most Comment Spam plugins resort to a simplistic classification of comment spam - spam or ham (genuine comment). Comment spam protection is more complex than what can be addressed with such simplistic schemes. As a result such plugins are forced to be either over-aggressive or over-defensive, mostly earlier. Additionally many (genuine) human generated comments are cleverly disguised to pass off advertisements as genuine comments by appearing to be mildly relevant. Such human generated comments often broach on a gray area of comment spam where blog authors debate on what is spam and what is not.
Comment Guard Pro uses advanced algorithms driven by expert system to uniquely classify comments in four distinct categories: Delete - Comment Guard Pro is absolutely certain that the comment is spam. As such (by default) it decides to delete them without further checks (saving processor cycles) or saving them as spam (saves database space as well as processor cycles).
Spam - Comment Guard Pro has high degree of confidence in the spamminess of comments and marks them as such, which moves them to Spam queue.
Moderate - Comment Guard Pro doesn't have enough evidence to classify the comment as either spam or ham. Very rarely comments are sent to moderation queue. As such the requirement for manual moderation is very minimal.
Approved / Ham - Comment Guard Pro has determined with a high degree of confidence that a comment is genuine and has decided to approve it.
To give an example, in Simple Thoughts blog over 1201725 (at the time of writing) comments were processed by Comment Guard Pro of which 98.21% were positively identified as spam and deleted, 1.1 % were marked as spam, 0.42 percent of the comments were approved and only 0.28 percent of the comments were marked for manual moderation.
Comment Guard Pro is designed with modular architecture from ground-up. Most of Comment Guard Pro functionality is encapsulated in pluglets, each of which independently checks a comment to determine its spamminess (spam, ham or requires manual moderation). Each pluglet is individually customizable, can be independently enabled or disabled and their execution order changed from the user interface using drag-and-drop.
Comment Guard Pro ships with 15 built-in pluglets. Additionally 7 pluglets are provided for free with Comment Guard Pro Power Pack. You can also develop (distribute and sell) your own pluglets.
Comment Guard pluglet API (open source, LGPL License) allows third-party developers like you to write your own pluglets to extend & enhance Comment Guard Pro's spam hunting capability and more. We provide the API along with detailed documentation and several examples to allow you to easily write your own pluglets.
You can provide your pluglet(s) to the community for free or you can even charge for it, we have no objection. We encourage free as well as commercial Comment Guard Pro pluglet development initiatives; your code, your choice.
Comment Guard Pro provides detailed live (and entertaining) statistics on comments, which is updated every 10 seconds (configurable), from options screen. You can see how effectively Comment Guard Pro is protecting your blog from spammers.
Comment Guard Pro provides comprehensive comment management facilities accessible from Manage->Comment Guard menu. You can now check the approved comments, comments marked for manual moderation as well as spam comments. Each queue is displayed in a separate tab and your choice of tab is remembered for easier access next time.
Each comment queue is displayed in multiple pages, if necessary, for easier viewing. You can search within comment text, author, e-mail, URL, IP address and reason; search results are paged.
For each comment you can:
Comment Guard Pro allows you to manually recheck the queue at any time, in addition to automated rechecking to protect you from sophisticated bait-and-switch comment spamming techniques described in the next section (Time Delayed Comment Rechecking).
Highly sophisticated comment spammers employ a bait-and-switch technique where they display tailored version of their content before spamming, so their comment passes through even the most advanced comment spam filters. However after verification, they revert back to their original page. This can defeat not only the most sophisticated comment spam filters but also humans.
Comment Guard Pro uniquely protects against such techniques with our innovative Recheck Comment pluglet. It rechecks each comment in approved queue (which hasn't been manually moderated) as well as moderation queue at least after a configurable number of hours (after posting) with a subset of rules (recheckable rules) to protect against such practices. No other comment spam filter in the market today has developed this technology.
You can also manually recheck the queue at any time from Comment Management Console.
Spammers sometimes use devious time-bound techniques to try to escape anti-spam filters like for example providing temporary link-backs to try to fool trackback link validator. Pluglets like Kismet (which uses Akismet API to identify spam) or DNSBL (which uses DNSBL services to identify spam) sometimes provide better judgement over time especially in cases when spammers decide to bombard large number of sites in a short period of time.
Recheck Comment pluglet provides each recheckable pluglets a second chance to correctly identify spams.