Slashdot Mirror
Ending Spam
Shalendra Chhabra writes "Jonathan
Zdziarski has been fighting spam since before the first MIT
spam conference in 2003, and has now released a full-on technical
book,
Ending Spam, on spam filtering. Ending Spam
covers how
the current
and near-future crop of heuristic and statistical filters actually work
under the hood, and how you can most effectively use such filters to
protect your inbox." Read on for the rest of Chhabra's review.
Ending Spam: Bayesian Content Filtering and the Art of Statistical Language Classification
author
Jonathan A. Zdziarski
pages
312
publisher
No Starch Press
rating
8
reviewer
Shalendra Chhabra
ISBN
1593270526
summary
Very Good Book Covering Statistical Models and Techniques Implemented in Current Spam Filters
Spam (unsolicited commercial email) and phishing (fraudulent emails) are causing losses of billions of dollars to businesses. Many initiatives are currently underway for fighting this challenge. On the legal front, a Virginia court recently sentenced a prolific spammer, Jeremy Jaynes, to nine years in prison, and a Nigerian court sentenced a woman to two and a half years for phishing. Michigan and Utah have both passed laws creating "do-not-contact" registries in July/August 2005, covering e-mail addresses, instant messaging addresses and telephone numbers. Technical initiatives to fight spam include server- or client-side spam filtering, using Lists (Blacklists, Whitelists, Greylists), Email Authentication Standards (IIM, DK, DKIM, SPF, SenderID), and emerging sender reputation and accreditation services.
Ending Spam is the first book explaining the fine details of the theoretical models and machine-learning algorithms implemented in these filters. The book is divided into three parts: introduction to spam filtering, fundamentals of statistical filtering, and advanced concepts of statistical filtering.
The first section of the book discusses the history of spam, spam kings, different approaches for fighting spam such as blacklisting, whitelisting, heuristic filtering, challenge response, throttling, collaborative filtering, Authenticated SMTP, Sender Policy Framework and SenderID, spammer fingerprinting, etc. However, the author omitted any mention of locally-sensitive hash functions (such as Nilsimsa Hash) to counter spammers' random insertion of words, the use of CAPTCHA (Completely Automated Public Turing Test to Tell Computers and Humans Apart), Greylisting, Identified Internet Mail, and Domain Keys (now Domain Keys Identified Mail).
In the next chapter, the author clearly explains various components of a Language Classifier Pipeline, including the Historical Dataset (aka wordlist, database, dictionary, filter memory), Tokenizer, and the Analysis Engine with its feedback loop. However, the process flow of a language classifier could have been more generalized, e.g. incorporating an initial text-to-text transformer. This chapter also covers the advantages and disadvantages of various training modes for filters, such as Train Everything (TEFT), Train-on-Error (TOE), and Train Until No Errors (TUNE). This part concludes with the description of Paul Graham's famous spam-filtering technique using Bayesian classification (as described in "A Plan for Spam"), Gary Robinson's Geometric Mean Test, Fisher-Robinsons Inverse Chi Square (including the source code for the inversion function), and some other tricks for optimizing spam- filtering accuracy.
The second part of this book deals with the fundamentals of statistical filtering. The author explains HTML and Base64 encoding, followed by a detailed description of tokenization techniques (e.g. Sparse Binary Polynomial Hashing). Then there's a discussion of the various tricks that spammers use for penetrating filters. Although these tactics are mentioned in John Graham-Cumming's "Spammers Compendium," Jonathan has very elegantly explained why some tricks work for spammers and some don't. This part concludes by addressing some of the resource, storage and scaling concerns raised by the large number of features generated from tokenization techniques.
The third part of this book deals with advanced concepts of statistical filtering. This includes the testing criteria for measuring accuracy of an email filter, and some advanced tokenization concepts, e.g. chained tokens (taking word-pairs and phrases into account, instead of individual words) generated using a sliding 5-byte window as mentioned in Sparse Binary Polynomial Hashing. The next chapter describes the Markovian Model implemented in the CRM114 Discriminator, but the author fails to describe different weighting schemes for features implemented in the Markovian-based version of CRM114. The author then describes the Bayesian Noise Reduction Technique for purging "out of context" data from the mail text. This chapter concludes with a very nice summary of collaborative algorithms and techniques, such as Message Innoculation, Streamlined Blackhole List, Fingerprinting, Automatic Whitelisting, URL Blacklisting, and Honeypot email addresses for snaring spammers' address harvesting bots.
The most interesting part of this book is the appendix, where the author presents interviews with John Graham-Cumming of POPFile, Brian Burton of SpamProbe, Marty Lamb of TarProxy, Bill Yerazunis of CRM114 Discriminator, and Jonathan Zdziarski of DSPAM (himself). I loved this section.
The salient points of the book: it's very easy to read; each chapter begins with a very thought-provoking introduction, and concludes with a crisp "final thoughts" section. The number of technical errors are very few in this print, and the illustrations are of good quality. Since the book is geared more toward the Bayesian and statistical generation of spam filters, the absence of certain spam-busting technologies is acceptable. However, a noticeable omission is the lack of discussion about measuring spam-filter accuracy, and what impact this has on setting filtration thresholds. A section on the economics of tradeoffs, and the use of a Receiver Operating Characteristic curve (ROC) would have been very helpful.
Overall, by putting together Ending Spam, Jonathan Zdziarski has made another significant contribution (after DSPAM) to the anti-spam community. Whether you are a system administrator, anti-spam researcher, engineer or a newbie interested in fighting spam, this book is a great reference.
William S Yerazunis and Richard Jowsey also contributed to this review. Shalendra Chhabra is a Graduate Student in Department of Computer Science and Engineering at University of California, Riverside. He is on the development team of CRM114 Discriminator and has presented his work at MIT Spam Conference 2005, Cisco Systems, and Stanford University. You can purchase Ending Spam: Bayesian Content Filtering and the Art of Statistical Language Classification from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
Spam (unsolicited commercial email) and phishing (fraudulent emails) are causing losses of billions of dollars to businesses. Many initiatives are currently underway for fighting this challenge. On the legal front, a Virginia court recently sentenced a prolific spammer, Jeremy Jaynes, to nine years in prison, and a Nigerian court sentenced a woman to two and a half years for phishing. Michigan and Utah have both passed laws creating "do-not-contact" registries in July/August 2005, covering e-mail addresses, instant messaging addresses and telephone numbers. Technical initiatives to fight spam include server- or client-side spam filtering, using Lists (Blacklists, Whitelists, Greylists), Email Authentication Standards (IIM, DK, DKIM, SPF, SenderID), and emerging sender reputation and accreditation services.
Ending Spam is the first book explaining the fine details of the theoretical models and machine-learning algorithms implemented in these filters. The book is divided into three parts: introduction to spam filtering, fundamentals of statistical filtering, and advanced concepts of statistical filtering.
The first section of the book discusses the history of spam, spam kings, different approaches for fighting spam such as blacklisting, whitelisting, heuristic filtering, challenge response, throttling, collaborative filtering, Authenticated SMTP, Sender Policy Framework and SenderID, spammer fingerprinting, etc. However, the author omitted any mention of locally-sensitive hash functions (such as Nilsimsa Hash) to counter spammers' random insertion of words, the use of CAPTCHA (Completely Automated Public Turing Test to Tell Computers and Humans Apart), Greylisting, Identified Internet Mail, and Domain Keys (now Domain Keys Identified Mail).
In the next chapter, the author clearly explains various components of a Language Classifier Pipeline, including the Historical Dataset (aka wordlist, database, dictionary, filter memory), Tokenizer, and the Analysis Engine with its feedback loop. However, the process flow of a language classifier could have been more generalized, e.g. incorporating an initial text-to-text transformer. This chapter also covers the advantages and disadvantages of various training modes for filters, such as Train Everything (TEFT), Train-on-Error (TOE), and Train Until No Errors (TUNE). This part concludes with the description of Paul Graham's famous spam-filtering technique using Bayesian classification (as described in "A Plan for Spam"), Gary Robinson's Geometric Mean Test, Fisher-Robinsons Inverse Chi Square (including the source code for the inversion function), and some other tricks for optimizing spam- filtering accuracy.
The second part of this book deals with the fundamentals of statistical filtering. The author explains HTML and Base64 encoding, followed by a detailed description of tokenization techniques (e.g. Sparse Binary Polynomial Hashing). Then there's a discussion of the various tricks that spammers use for penetrating filters. Although these tactics are mentioned in John Graham-Cumming's "Spammers Compendium," Jonathan has very elegantly explained why some tricks work for spammers and some don't. This part concludes by addressing some of the resource, storage and scaling concerns raised by the large number of features generated from tokenization techniques.
The third part of this book deals with advanced concepts of statistical filtering. This includes the testing criteria for measuring accuracy of an email filter, and some advanced tokenization concepts, e.g. chained tokens (taking word-pairs and phrases into account, instead of individual words) generated using a sliding 5-byte window as mentioned in Sparse Binary Polynomial Hashing. The next chapter describes the Markovian Model implemented in the CRM114 Discriminator, but the author fails to describe different weighting schemes for features implemented in the Markovian-based version of CRM114. The author then describes the Bayesian Noise Reduction Technique for purging "out of context" data from the mail text. This chapter concludes with a very nice summary of collaborative algorithms and techniques, such as Message Innoculation, Streamlined Blackhole List, Fingerprinting, Automatic Whitelisting, URL Blacklisting, and Honeypot email addresses for snaring spammers' address harvesting bots.
The most interesting part of this book is the appendix, where the author presents interviews with John Graham-Cumming of POPFile, Brian Burton of SpamProbe, Marty Lamb of TarProxy, Bill Yerazunis of CRM114 Discriminator, and Jonathan Zdziarski of DSPAM (himself). I loved this section.
The salient points of the book: it's very easy to read; each chapter begins with a very thought-provoking introduction, and concludes with a crisp "final thoughts" section. The number of technical errors are very few in this print, and the illustrations are of good quality. Since the book is geared more toward the Bayesian and statistical generation of spam filters, the absence of certain spam-busting technologies is acceptable. However, a noticeable omission is the lack of discussion about measuring spam-filter accuracy, and what impact this has on setting filtration thresholds. A section on the economics of tradeoffs, and the use of a Receiver Operating Characteristic curve (ROC) would have been very helpful.
Overall, by putting together Ending Spam, Jonathan Zdziarski has made another significant contribution (after DSPAM) to the anti-spam community. Whether you are a system administrator, anti-spam researcher, engineer or a newbie interested in fighting spam, this book is a great reference.
William S Yerazunis and Richard Jowsey also contributed to this review. Shalendra Chhabra is a Graduate Student in Department of Computer Science and Engineering at University of California, Riverside. He is on the development team of CRM114 Discriminator and has presented his work at MIT Spam Conference 2005, Cisco Systems, and Stanford University. You can purchase Ending Spam: Bayesian Content Filtering and the Art of Statistical Language Classification from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
While all of these different technological approaches to spam are worth pursuing, they just don't build the same esprit d'corps as a mob with pitchforks and torches at midnight.
If brevity is the soul of wit, then how does one explain Twitter?
is with a knife, a spatula, and a frying pan, preferably over a hot wood fire.
Yum!
-- Tigger warning: This post may contain tiggers! --
From:
HERE
"ABOUT THE AUTHOR:
Jonathan A. Zdziarski has been fighting spam for eight years, and has spent a significant portion of the past two years working on the next generation spam filter DSPAM. His research in algorithmic theory and neural networking has led to the development of many new approaches in language classification, and he has played a key role in designing some popular algorithms in use today, including Message Inoculation, Bayesian Noise Reduction, and the first functional Neural Networking algorithm for spam filters. Zdziarski lectures widely on the topic of spam and was a speaker at the 2004 and 2005 MIT Spam Conference.
"
You can't talk about Wikipedia's flaws on Wikipedia
If you can't see it, it ain't there?
I'm wondering... will UCE (Spam) be like malaria... controllable in most areas but impossible to eradicate?
Or will these dedicated folks and others be able to eliminate it, perhaps by changes to the mail protocols?
Interesting question that, considering my work involves malaria.
My guess is that, like malaria and most parasitic infestations, we will at some point develop a "cure". The "cure" will work for a few years, after which the parasite (spam) will have adapted, surviving until then in different hosts (old windows machines donated to Africa, who knows). Then, having developed a new trick, it will come back as strong as ever.
Biology teaches us that organisms adapt to changing environments, thru selective breeding (natural), point mutations, and unforseen combinations (see the H51N avian influenza). We can develop cures, but once we do so, we can be fairly sure that, baring species extinction, it will develop methods to cope with our cures.
An easy solution would be to move to IPv6 - but this, like authentication, will only kill off the spam which doesn't use "trusted email clients that are identified" while the spam that can survive will be encouraged to spread like wildfire.
So long as the fiscal, legal, and societal penalties for spamming are fairly low and the rewards are high, and while most people do nothing about it, it will spread.
-- Tigger warning: This post may contain tiggers! --
Email, as a system, is fundamentally broken. It's this broken design that allows SPAM to happen in the first place.
Current anti-spam solutions are to email what an Antivirus package is to Windows - a hack add-on that increases complexity and costs without solving the underlying problem(s).
Rather than fight viruses, we should be engineering an O/S that's inherently resistent to them. How many of you Linux/BSD/MacOS users EVER use antivirus, or need to?
Rather than build ever-better antispam filters for Email, we should be engineering an email solution that's inherenly resistant to SPAM.
The answer lies in authentication - who is sending the email. Some of the best technologies now available use degrees of authentication without actually *saying* it outright. Examples are: refusing invalid domains, greylisting, challenge-response, SenderID - all of these are some form of authentication.
As these are, one-by-one bypassed by the spammers, the need for authentication of senders will continue to increase, until the dolts who will invariably reply with that "your solution will not work because... (check the options)" are shown to simply be.... wrong.
Give it time. It's already happening whatever the originators of the SMTP protocol desired.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
The reason spammers do it is that their message reaches people, enough of them to make it worthwhile. So, the more effective and widespread the filters, the less messages that reach people, and the less it's worth. If the filters were really effective, nearly 100%, it would simply not be worth it to spam, you wouldn't make any money because no one would see your message.
I don't think we'll ever get there, but yes filtering really could end spam.
This may be the case; however, that doesn't invalidate his work on spam. Remember, Sir Isaac Newton was a firm believer in the more exotic aspects of mystical alchemy, and the vast bulk of his 'research' was complete gibberish. That doesn't make his work on gravity any less valuable.
Spam is no longer simply the domain of a giant server with a huge database. It's increasingly being sent out by zombie PCs, infected with viruses or trojans. Spammers pay the zombie-farmers to send their crap. Zombies send the email masquerading as the PC owner, using their credentials. Sender-ID? No problem, he's got one. SMTP? Sure, use the victim's server.
Zombies mean that no matter what technology is used for sending validated, signed, pre-paid, whatever email, the zombies will have access to those resources and will still spew their crap. No anti-spam server technologies are going to prevent Windows machines from getting infested.
John
The first step to a new mail system is to assure that only legitimate and properly configured mail servers honoring MX records on outgoing mail (or whatever ends up replacing MX records) can expect delivery. Mail admins' hands are tied by stealth systems or badly configured ones, but if we do try to implement the no-MX rule, which would eliminate the zombie attacks, we end up shutting out systems that, for whatever reason, don't publish an MX record for outgoing servers.
Zombies ought to be the easiest thing to shut down by a) not permitting non-MTA machines to push anything beyond the network via port 25 and b) publishing both incoming and outgoing mail servers.
The world's burning. Moped Jesus spotted on I50. Details at 11.