At 05:12 AM 8/13/2003, Brad Knowles wrote:
I think we were planning on looking at these specific characteristics when analyzing spam data. Paul Judge (the chair) was working on that, you can contact him for more information.At 10:16 PM -0400 2003/08/12, Yakov Shafranovich wrote:I've seen that before. While it is a list of interesting characteristics that tend to be common with spam, it has not been clear to me how this relates to the overall effort.Can you give us any guidelines as to how we might be able to help analyze this data?First of all take a look at the following link which lists various characteristics to look out for: http://www.irtf.org/asrg/spam_characterization.htm
The group is gigantic and there are eight major areas of research. We would like to see each of those eight areas (and possibly some of the large sub-areas) to be headed by someone. Heading an area would imply coordinating efforts in that area, getting people to volunteer, etc. In the analysis area specifically it would not hurt to know statistics, but I think that it is not necessary. If you are willing to manage that area, you can get someone to help who does know statistics......In charge? In what way? Does this person need to be deeply knowledgeable in statistics and applying statistical methods to the analysis, or would that work be done by someone else?Third, we need someone to be in charge of the analysis work, coordinating various research in that area.