Spam and Personal Privacy

1
Spam and Personal Privacy

• Presented by Ashley Embry

2
(No Transcript)
3
Outline

• What is Spam?
• A. Types of Spam
• Where Did the Word spam Originate?
• How Spam Begins A General Explanation
• Who Has the Potential to be a Spammer?
• Statistics About Spam
• Getting Rid of Spam
• Breakdown of a Spam Filter
• Conclusions
• Questions for the class

4
What is Spam?

• There are many definitions of spam that are used.
• Electronic junk mail or junk newsgroup postings.
• Any unsolicited automated e-mail.
• Email advertising for some product sent to a
  mailing list or newsgroup.
• Spam is simply flooding the internet with many
  copies of the same message in an attempt to force
  the message on people who would not otherwise
  choose to receive it.

5
Types of Spam

• There are two main types of Spam
• 1. Usenet Spam is aimed at people who read
  newsgroups but rarely or never post and give
  their information away.
• 2. E-mail spam targets individual users with
  direct mail messages. E-mail spam lists are
  created by scanning Usenet postings, stealing
  Internet mailing list, or searching for
  addresses.

6
Where Did the Word spam Originate?

• The history of calling inappropriate postings in
  great numbers spam is from a Monty Python skit
  where a couple goes into a restaurant and the
  wife tries to get something other than Spam. In
  the background there is a group of Vikings who
  are singing the praises of Spam. Pretty soon the
  only thing that you can hear is
• Like the song spam is the endless repetition of
  worthless text.

7

• Another proposal is that spam was thought of by
  a computer lab group at the University of
  Southern California, who gave it the name because
  it has many of the same characteristics as the
  lunch meat Spam.
• Nobody wants it or ever asks for it.
• No one ever eats it it is the first item to be
  pushed to the side when eating the entrée.
• Sometimes it is actually tasty, like the 1 of
  junk mail that is really useful to some people.

8
How Spam Begins A General Explanation

• Spammers only need access to your address. After
  that its just a matter of sending the e-mails.
• The primary sources that spammers use are
  newsgroups and chat rooms.
• The second source used is the Web itself.
  Spammers can create search engines that look for
  the _at_ sign which indicates an e-mail
  address.
• The third source is sites created specifically to
  attract e-mail recipients.
• Win 1 million!!! Just Click Here!
• Would you like news letters form our partners

9

• Finally, probably the most common source of
  e-mail addresses comes from searching the e-mail
  servers of large e-mail hosting companies like
  Hotmail.
• The Hotmail article A Spammers Paradise reads

A dictionary attack utilizes
software that opens a connection to the
mail server and rapidly submits millions of
random e-mail addresses. Many of these
addresses have slight variations, such as
"jdoe1abc_at_hotmail.com" and jdoe2def_at_hotmail.com.
The software then records the address
locations and adds those

addresses to the spammer's list.
These lists are typically resold to

many other spammers .
10
Who Has the Potential to be a Spammer ?

• Anyone can be a spammer.
• Scenario
• Lets say your grandmother bakes the best banana
  nut bread ever created, and you want to sell the
  recipe for 5.
• You have 100 people in your personal e-mail
  address book. You send out an e-mail advertising,
• Big Mommas Nana Nut Bread - only 5 !!!
• From your 100 e-mails you get 2 orders and make
  10.
• Imagine if you had sent out 1,000,000 e-mails

11
Statistics About Spam

• In a single day in May, the No. 1 internet
  service provider AOL Time Warner (AOL) blocked 2
  billion spam messages88 per subscriberfrom
  hitting its customers e-mail accounts.
• Microsoft (MSFT) which operates the No.2 service
  provider MSN and Hotmail says it blocks an
  average of 2.4 billion spams per day.

12
Getting Rid of Spam

• Avoid giving out your e-mail address to
  unfamiliar or unknown recipients.
• Use your e-mail applications filtering features.
• Report the spam e-mailer to the spammers ISP.
• Use spam filtering software.

13
Breakdown of a spam filter

• Most spam blockers use filters that search for
  commonly used phrases or writing styles that are
  overly aggressive and found in mass e-mail
  marketing. Spammers try to fool the filters by
  changing their writing styles and formats so that
  their messages can sneak past the filters.
• The best technology currently available to stop
  spam is spam filtering software.
• The simplest filters use keywords such as xxx,
  viagra, etc, but they are also more likely to
  block the e-mails that you do want to receive.

14
Example

• The more advanced filters, Bayesian filters for
  example, take this approach further to
  statistically identify spam based on frequency.
• An example of how this statistical filtering
  works
• Start with one collection of spam and one of
  nonspam mail, and each collection had about 4000
  messages in it.
• Scan the entire text of each message of the
  collection.
• Consider alphanumeric characters, dashes,
  apostrophes, and dollar signs to be as part of
  tokens (words) and everything else to be a token
  separator. (i.e. qt234abc, 75, utt)
• Count the number of times each token occurs in
  each message. You will end up with two large
  tables with each one showing the different tokens
  and how many times it appeared in the messages.

15

• Finally, create a third table that relates the
  token to the probability (ranging from .01 to
  .99) that an e-mail containing it is a spam.
• When new mail arrives now, it is scanned into
  tokens, and the fifteen tokens whose
  probabilities are the farthest from the neutral
  probability of .5 are then used to calculate the
  probability that the e-mail is a spam.

16

• Algorithms/Program language
• To determine probability of the token being in a
  spam
• let ((g ( 2 (or (gettable token good) 0 ))
• (b (or (gettable token bad) 0 ))
• (unless (
• (max .01
• (min .99 (float (/
  (min 1 (/ b nbad))
• ( (min 1 (/ g
  ngood))
• (min 1 (/
  b nbad))))))
• To determine if the e-mail is a spam using the
  probabilities of the 15 chosen tokens
• let ((prod (apply probs)))
• (/ prod ( prod (apply (mapcar
  (lambda (x)
• 
  (-1 x))
• 
  probs))))

17

• Example token list with probabilities
• madam 0.99
• promotion 0.99
• shortest 0.047225013
• sorry 0.0499
• valuable 0.82347
• information taken from www.paulgraham.com

18
Wrapping it Up

• Whether constructing a spam list or implementing
  a spam filtering program, spam is based on the
  concept and utilization of computer science.

19
Questions for the Class

• By the end of this presentation you should be
  able to answer the following question
• Name 2 techniques we learned in CIS class that
  are used by spammers or in spam filtering.
• Pattern-Matching when searching for email
  addresses or when evaluating words for spam
  tendencies.
• Writing algorithms to eventually implement
  program.

20
Bibliography

• Before Spam Brings the Web to Its Knees. June
  10, 2003.
• http.//www.businessweek.com/technology/content/jun
  2003/tc20030610_1670_tc104.htm
• Brain, Marshall. How Spam Works
• http//computer.howstuffworks.com/spam.htm
• Getting Rid of Spam
• http//www.webopedia.com/DidYouKnow/Internet/2002/
  GettingRidofSpam.asp
• Graham, Paul. A Plan for Spam. Aug.2002.
• http//www.paulgraham.com/spam.html

21

• Mueller, Scott H. What is Spam?
• http//spam.abuse.net/overview/whatisspam.shtml
• Origins of Spam
• http//digital.net/gandalf/spamfaq.htmlitem8c
• Spam July 20, 2004.
• http//www.webopedia.com/TERM/s/spam.html