Reflective Reality 3
Section 2 on learning based filters (Last session)
Batch and Online Spam Filter Comparison; Gordon Cormack and Andrej Bratko. Gordon offers a plug for Trec ;)  Read the paper, I can’t summarise this one! Captivating presentation by Gordon on the good bad and ugly comparison of DMC PPM Bogofilter LR and SVN classifiers. Possibly the best [...]
Teaching Spam and Spyware at the University of C@1g4ry; John Aycock – I had coffe with John yesterday, and he has a strong personal belief that uni’s can help scociety by educating CS students in fighting spam and spyware . He teaches a full a 13 week course at [...]
Modeling Identity in Archival Collections of Email: A Preliminary Study; Tamer Elsayed, Douglas Oard. More nodal analysys of the Emron corpus using address-name/nickname-address associations in headers and quoted headers. 96.7% entity accuracy is pretty impressive especialy with the weakest evidance. I have to wonder what else this could be used for if they [...]
Day 2.
Keynote: Hector Garcia-Molina and Zoltan Gyongyi, Stanford – Web Spam: Current Techniques and Future Challenges. A graphical view of search engine pollution and techniques for boosting pages content and links. (Repetition, dumping, weaving, and stitching) Classic research stuff. Zoltan has been working with yahoo on Trustrank, demoting pages with high pagerank based on [...]
1 Eilon Solan (former Blue Security) – Charging for mail & another registry. Frankly I’m not going there on this one. Both topics are not new. Nice cost modeling though Lets move on.
2 Brad Taylor – Google. Gmail’s ‘reputation’. RBL’s are old eh ? I wonder who they have feeds from . The [...]
Section on : Email Enhancement
1 An Email and Meeting Assistant Using Graph Walks; Einat Minkov, William Cohen. Interesting research presented by Einat into concurrency relationships between extracted logical graph points. An using lazy grap walks to form relationships between messages. Nice to see the string matching comparisons too.
2 Email Thread Reassembly Using Similarity Matching; Jen-Yuan [...]
Section on : Counter-Attack
1 : Spamalot: A Toolkit for Consuming Spammers’ Resources; Peter Nelson, Kenneth Dallmeyer, Lucasz Szybalski, Tom Palarz, Michael Wieher – A very fun presentation regarding using an AI system to use up and abuse the 419ers and loan spammers time with a 65% success rate and a average chain length of email [...]
Section on : Abuse detection and control
Presentation 2 is on Dynamic Port 25 Blocking to Control SPAM Zombies by Jonathan Schmidt.
In their experience dynamic blocking is almost as as effective as a total port 25 block this is since many end users don’t actually care about port 25 blocking. They had 5 out of 250,000 [...]
So presentation 1is “Wacky” Rob Thomas from team Cyrmu. Is a great presenter on the underground economics of networks of spam/phish. The web of the underground in an entertaining and pictorial form. Nothing really new from my point of view, though the router section is bigger than I had imagined.
“If you thinking the changing the [...]
I’m off to CEAS 2006 later in the week. The weather is looking much the same as home so with any luck I’ll not have to stow luggage
The agenda is looking pretty interesting this year with a nice balance of topics evenly spread over the two days.
I’m staying in the conference hotel so [...]
