Spam Liquidator: an Innovative Self-Learning Spam Filtering Software

Spam Liquidator's Anti-Spam Filter is an innovative self-learning system offering you spam protection without deleting the important correspondence. Anti-Spam Filter maintains its own collaborative spam-tracking database, which uses the signatures of spam messages. It is different from other anti-spam programs, in the way that you're not the only one to "teach" it, but all of our project users contribute to it.

Upon studying many spam-detection algorithms, we have come to a conclusion that only a human being is truly a hundred percent capable of telling spam from regular messages. Even the most complex algorithms can make mistakes, and can either pass the spam through or, much worse, delete an important letter. The only way of generating a real spam killer is uniting all of our efforts.

How does it work? Aside from initial spam filters, provided by our program and created by our developers, our Anti-Spam Filter presents a claim submission practice, in case you have got some spam coming through. The process is ultimately simplified - to report spam you just have to make two mouse-clicks.

The development of the Spam Liquidator's Anti-Spam Filter has begun in September 2002. As of February 20 2003, the project has been open to public via Internet. Program functions are regularly being updated and supplemented. We are open for all suggestions and propositions.

Join Spam Liquidator to make spam history!

Spam Liquidator guarantees stopping spam without deleting the important letters with the use of its Anti-Spam Filter software.

Spam is a growing problem for regular Internet users and business organizations. It reduces employees' productivity, puts your computer at risk of getting numerous viruses, wastes your time and simply annoys you.

Features

The unique aspect of Spam Liquidator's Anti-Spam Filter is our claim submission practice, which the principles of spam blocking are based on. It is an automatic process that provides exceptional operational efficiency of your system.

As all other derelicts, spammers have some kind of communities. They have the same distribution and fraud methods, same address databases of their victims. Moreover, upon studying statistics, we have concluded that aren't that many databases and they contain pretty much the same addresses. Therefore, there is a 50% guarantee that you and I will receive the same letter.

Let's consider that a spammer sent out a letter to a million people. Half of the addresses are going to be invalid or forgotten. But half a million is still a lot. Even if the letter is no more then 3kb, it's 1,5GB of wasted traffic! However, when you or I, or any other user receives this letter, it can be reported and after several complaints, the letter will be blocked among all Anti-Spam Filter users.

So, what do we have? Say, a spammer has been sending out letters the whole night. Early morning it will be reported by several users, the rest will just receive a notice that the letter has been deleted. If the initial number of users is a thousand people, only ten will receive this letter, which makes it 10% and guarantees you 99% spam protection without the risk of deleting important correspondence.

Some of you may wonder if your friends will play a prank on you. Don't worry about it. One complaint is not enough for the system to block your e-mail address. It is also useless to repeat the prank complaint, because a user can only send one complaint for each spam address. The repeated complaints will be ignored. For you to get blocked, you need to be a spammer yourself.

Anti-Spam Filter also uses other ways of identifying spam, such as:

The filter can recognize a number of tricks, which spammers use to mask their identities, fool you into thinking they've sent a valid mail, or fool you into thinking you must have subscribed at some stage.

Again, spam mails often have a characteristic style, some characteristic disclaimers and CYA text. Anti-Spam Filter will identify these as well.

Anti-Spam Filter supports many existing abuse blacklists, such as DSBL (Distributed Server Boycott List), DNSBL (Domain Name System Black List), DNSRBL (Domain Name System Real-time Black List), ORDB (Open Relay and Open Proxy Database), and many others. Its filters do not rely solely on the abuse blacklists because that would lead to high false positive rates. Instead, we review each spam claim contributed by our project's users, and, if the spam sample is sent from the IP address that is on one of the existing spam lists, we give such report higher priority.