Slashdot Mirror
Using AI for Spam Filtering (w/ Source Code)
jarhead4067 writes "Article snippet: "Up until recently, most researchers in the fight against spam have failed to classify it as an artificial living organism, hindering the development of effective tools and techniques to kill it. While this classification may sound strange, consider the following..." A novel approach to filtering spam, and hey, there's free source included."
... after we get an AI to counter the Slashdot effect.
"Enough of this wretched, whining monkey life." -- Marcus Aurelius, _Meditations_, Book 9, 37
Google cache
I won't believe spam is a living organism till I see Marty Stouffer do a special, complete with comedy 'boing' noises and 'aint that cute' music as we watch a mother Spam care for her young.
"living organism ... and techniques to kill it"
Next thing we know, we will have Animal Rights Activists in Washington, D.C. protesting our "spam traps"
who | grep -i blond | date cd ~; unzip; touch; strip; finger; mount; gasp; yes; uptime; umount; sleep
I mean - hello, humans create it.
We're not up against a new being - it's the same type of beings that create scripts for the hell of it that wreak havoc on computer networks because 1) "We can" or 2) "To show them their weaknesses".
It was a very interesting read for sure - the genetic marker bit was quite interesting. Admittedly though I got about 2/3rds the way through it and lost interest.
Blame the spammers I say. ^_^
Spam has become the first great plague of the 21st century. Over 60% of all e-mails are spam, costing U.S. corporations more than $10 billion annually, on top of the productivity lost from scanning through e-mail and deleting spam. Along with this, an estimated 5% of spam campaigns are a pure and outright scam, with the remaining majority pitching products that are dubious at best. It used to be parents had to worry about their kids surfing and finding pornographic websites, now we have to worry more about our kids opening an e-mail client and finding a pornographic spam message. Spam must be stopped before it cripples the infrastructure of the internet and drives users away from one of the greatest forms of communication, E-mail.
Can Laws Defeat Spam? No. This has to be one of the greatest misconceptions of users. The internet is just that, an "INTERnational NETwork" that cannot be governed by one country's laws. Spammers can exist anywhere on the internet, meaning they can sling their wares from anywhere in the world, making the laws of one country completely irrelevant. Also, the decentralized, self-organizing design of the internet makes it nearly impossible to regulate by external means. It would be easier to regulate the weather than to regulate the internet.
Spam as a Living Organism
Up until recently, most researchers in the fight against spam have failed to classify it as an artificial living organism, hindering the development of effective tools and techniques to kill it. While this classification may sound strange, consider the following:
Through the fight against spam, spam has demonstrated an uncanny ability to adapt to the conditions of its environment, namely the internet. When one barrier against a strain of spam is put up, another, resistant strain appears. This is similar to how bacteria builds immunity against antibiotics, the strains that are not immune will die, while the ones that are immune take over and become the dominant, drug resistant strain. This leads to the belief that spam will not die until the barriers of its environment evolve faster than it does.
The internet is a complex chain of systems that all rely on each for the other's survival. Without an internet protocol, a web browser couldn't exist. Without web servers, the web wouldn't exist. Without ... (you get the picture). This chain of systems can be likened to an eco-system, with spam existing at a parasitic level of species within this system. It consumes resources (bandwidth, servers, time) in its attempt to reach its primary host: us. Once spam reaches its target, its sole purpose is to solicit its "food" from us, primarily money. If it is effective, that strain of spam lives and continues to propagate, otherwise it will die. Can the internet eco-system be modified so spam can't feed?
Just like any organism, spam contains certain traits that uniquely identify it. This can be a combination of words, information inside the header of the e-mail, the format of the message (HTML, plain text, rtf), the message encoding (base64), does it contain image links, the number of links, does it contain hidden text, so on and so forth. Up until recently, spam filters have primarily focused on just one of these traits, the wording of the e-mail. Spam, being an organism, evolved so this marker was hidden within its code, making it difficult at best to filter. It did this by including random, non-spam words in hidden areas of the e-mail, by modifying words like Viagra with V1@gr@, sending spam as image links, and by encoding the message in a format that filters could not read. The good news is this "gene" is still present, and can be unlocked by identifying the defensive genes wi
wot no sig
How exactly is this news ? It seems that the author of the neural network idea didn't do his homework - e.g. DSPAM includes neural network as an experimental classifier already. And compared to the proposed C# solution, DSPAM is a widely used and mature product already.
Regards, Jan
We already saw a plagiarized article green-lighted, and now this? Cmdr Taco, Slashdot was a brilliant idea of yours, and I love your site -- but that's because I have reasonably high expectations for it.
First, the submitter of this article has he email address jarhead4067@hotmail.com -- and so does the article's author.
Second, what is presented is not a genetic algorithm. The characteristics of the email to be considered to discover if the email is spam are finite and hard-core -- and even the threshold some characteristics must reach to qualify as spam are hard-core:
A genetic algorithm is one in which the goal is hard-core, different means of reaching that goal are generated, and the characteristics of the most successful are used to generate the next "generation"; this is repeated until the goal is reached.
But in this model, each "chromosome" contains statistics about one email. The heart of this model is to train a neural network with known emails ("chromosomes") and then tests unknown emails ("chromosomes") against the network.
Neural networks have a checkered history in Artificial Intelligence research. A (very much simplified) model of biologic neurons, neural networks were for a time seen as a great hope for Artificial Intelligence. A neural network basically starts out with an array of input nodes and an array of output nodes, with each input node connected to each output. Each input corresponds to some characteristic of the items the network is trained with: for classifying animals, the inputs would be characteristic of animals, e.g., "furry", "bipedal", "feathered"; each output a classification, e.g., "mammal", "bird", "human".
To train the network, the input nodes are set to the characteristics of an item, and then the strength of the connection of those inputs to the correct outputs is increased (or that of other connections is decreased -- it's the same thing). With enough training, it's possible to isolate the salient characteristics from the ambiguous one sin a mechanistic way.
This is useful, but it was soon discovered that these simple neural networks, for certain sets of inputs, failed, because of overlapping categories: both birds and humans are bipedal, but only humans are also mammals. In a single layer neural network, the connection strength between input "bipedal" and output "mammal" would fluctuate, unable to describe humans or birds well. These problems can be alleviated by adding additional "hidden" layers of nodes between input and outputs, and by allowing "back-propagation" from output or hidden nodes to layers "previous" to them.
But even with these enhancements, it's been conclusively shown that some problems are intractable for neural networks. In any case, neural networks are no new thing.
Of course I have no idea if classifying spam is intractable or not, but I have to question whether using a neural network reliably can outperform Bayesian (or quasi-Bayesian) filtering. My guess is that since Bayesian filtering can judge email by the occurrence of single tokens ("words"), and not just "chromosome" statistics, and given that this "new" method also uses Bayesian filtering to generate one of those "chromosome" statistics anyway (and for only the most difficult to characterize emails to boot), this method itself probably mostly relies on its Bayesian sub-component.
So I'm a bit at a loss to see why this method is in any way revolutionary or even particularly interesting, or why it was green-lighted for Slashdot. Of course, I only gave the linke
Opinions on the Twiddler2 hand-held keyboard?
from SpamAssassin? It takes a bunch of rules, applies them, and uses a neural net to classify the message. Seems to me SpamAssassin does the same thing, only is more mature and extensible and uses a genetic algorithm rather than a back-propagation neural net.
I think about the only good thing I can say about this article is, at least he's not out killing puppies.
Religion is a gateway psychosis. -- Dave Foley
I've given up on Spam filtering and concentrating my efforts on Ham filtering.
Basically the present thinking is based on attempting to filter spam out - I would argue that given the amount of variables involved, it it a method doomed to failure. Current methods also assume that the incoming mail is mostly valid, and are attempting to remove the undesirable parts - spam.
What I am having success with is turning this on it's head and assuming that the bulk of incoming mail is bad, and filtering in messages that I want.
The way I am doing this is to use my address book as a whitelist - if an incoming message originates from someone in my address book, then it's delivered into the inbox. If not, then they are moved into a "not in address book" sub folder. Anything my ISP spam assassin based filtering marks, is sent into the "Spam" folder. Doing it this way means that I am only notified of incoming mail that is confirmed from someone in my address book. Periodically I check the other folders (obviously).
We have come to the point I think where the number of variables involved makes filtering in a less intensive process than attempting to deal with the myriad of underhanded techniques that spammers use. By limiting the mail I want to people in my address book, I make it so that spammers are the ones having to deal with the variables as they would have to guess addresses in my address book. If lots of people started filtering like this when we would see spammers using known bulk mail addresses (such as the address iTunes receipts are mailed from) however we can simply alter the filter to include the originating IP / mailer and so on.
Think of it like fishing - you wouldn't attempt to control an entire ocean and remove the water to leave the fish - you accept that the water is there and develop techniques to get the fish out.
1. While the author proposes some marvelous cure based on treating spam as an organism, he just lists traits that any spam filter can use, and which most probably do, though he would suggest that most don't. I fail to see how the artificial-life observation improves spam non-spam determination from the list of traits he proposes filtering on.
2. The article reads like a sales pitch for the author's spam filter.
3. If 2 is true, and it is a sales pitch, then you have the irony of a very effect form of spam that makes it past the slashdot editors.
It's ALIVE!!!!
Letter To Iran
If I were to sum up this approach, it would be SpamAssassin with a multi-layer neural network. I should mention that I maintain the tool that SpamAssassin is useing to train its single-layer neural network for version 3.0, so I can honestly say that have a fair amount of experience in this area.
I'm not too keen on Evans' use of the biological metaphors. I think that they only confuse the issue of what he is doing. I will use the standard terminology, features, from here on out.
What he is doing is finding a nonlinear decision surface between two classes using a universal function approximator. I will explain this in layman's terms.
Imagine a sheet of paper filled with multi-coloured dots where these dots are arranged in clusters and each cluster contains mostly the same number of dots. Starting with a simple example, imagine two clusters of dots, one blue and one red. Assume that you can draw a line that separates the two clusters. That line is called the decision surface. You would say that any new dot that would appear on one side of the line will be called red and the other blue. Any blue dot that appears on the red side of the line would be misclassified as red. This is referred to as a linearly separable problem.
Now, imagine a more complex arrangement of clusters where you can't draw a straight line to separate the red from the blue, but you can separate them using a curved line. This is called a nonlinearly separable problem.
Artificial neural networks are very good for representing these decision surfaces. They are constructed of one or more perceptrons. A perceptron uses an activation function and a transfer function to take a set of inputs and produce a single output. The most popular form of neuron uses a linear activation function and a sigmoid transfer function. The linear activation function is the sum of a set of weighted inputs, i.e. f(X) = sum w_i *x_i. The logarithmic sigmoid transfer function is g(x) = 1/(1+exp(-x)). The output of the perceptron for any given input is O(X) = g(f(x)).
These perceptrons can be chained together in many different ways. One popular method is the multi-layer perceptron, where a set of neurons in the hidden layer process the inputs and pass on their outputs to the output layer where the final output is formed. I don't have a source for you, but it has been proven that, given a large enough hidden layer, the multi-layer perceptron is a universal function approximator.
As long as all of the transfer functions are differentiable, you can train a neural network using error backpropagation by gradient descent. I will leave it as an exercise to the reader to learn how it works, but I assure you that it is very simple. Machine Learning by Tom Mitchell has a good section on the subject, as does Fundamentals of Computational Neuroscience by Thomas Trappenberg.
Evans has identified a large set of features of e-mails, some of whom on their own convey little or no information about whether an e-mail is spam. He trains the neural network to recognize the combinations of these features which can lead towards the conclusion that a message is or is not spam. While his approach is a good idea, I would hesitate to call it novel. Massey, Thomure, Budrevich and Long did a very similar experiment [3] where they used a multi-layer neural network with SpamAssassin.
While his approach is good, there are some downsides for widespread deployment that need to be addressed first. With a large feature set like he is using, you will probably need a lot of training data to find a good fit with a multi-layer perceptron. To train the single layer neural network for SpamAssassin 3.0, I'm using 160000 messages.
Also, as his own arguments show, spam adapts to spam filter technology. Most of the features that he presents in his whitepaper can be easily fooled by a spammer. They can deliberately manipulate these features to evade the spam filter b
Worse than being killed by the AI.. what if the AI decides to not filter spam anymore?
"I'm sorry Dave, but your wife thinks you SHOULD try this V@GR!A substance."
or
"This Nigerian seems very nice, and if it pays off you can get me more delicious RAM."
Learn something new.