Spam Filter
Spam blocker
Spam According to Return Path Inc.'s "Email Blocking and
Filtering Report," for the first half of 2003, 17 percent of
permission-based e-mail messages weren't delivered to recipients by the
top ISPs. The e-mails were mistakenly tagged as spam. As ISPs and
consumers use tougher spam filters, marketers who work hard to get
permission to send messages will unfortunately see their response rates
dwindle.
Marketers can help protect their e-mails from being inadvertently
blocked. Permission-based e-mail marketing company e-Mail Networks Inc.
offers a few tips to improve delivery success:
1. Watch the headers. Avoid sending e-mail to a large number of
addresses in the "To", "CC" or "BCC" fields. If you use an e-mail
system, make sure it sends one message at a time.
2. Craft subject Lines with care. Write a subject line that shows it's
from a trusted source. Avoid using words and symbols that can appear to
be spam, such as "free," words in all caps or exclamation points.
3. Respond to e-mail requests. Instead of clicking an "unsubscribe"
link, which should be included in your routine e-mail communications,
subscribers may personally ask you to remove their addresses from your
database. Do this immediately, and report the resolution to the
recipients. Responding to their requests helps avoid a potential spam
complaint, which could put you on ISPs' block lists.
4. Get added to "white lists." ISPs such as AOL and EarthLink allow
subscribers to add e-mail addresses to their personal white lists. These
lists specify the e-mail addresses allowed to pass through the ISP's
spam filters. On your Web site page where visitors subscribe to e-mail
communication from you, ask them to add your address to their white
lists.
You may also want to include instructions for Internet users who are now
using spam filters within their own e-mail programs, such as Microsoft
Outlook. Users can add your company's e-mail address or domain name to
their "safe senders" list.
5. Monitor your bounce rate. When specific e-mail addresses bounce back
repeatedly, you can personally contact those individuals and ask to be
added to their personal white lists. If you don't get a response, remove
those e-mall addresses from your database. If you use an e-mail
management system, find out if invalid e-mail addresses are removed for
you. Too many bounced e-mails can trigger spam filters.
Also remember to post your privacy policy and unsubscribe instructions
on your e-mail sign-up page. These assurances soothe your Web site
visitors' concerns about getting secondary spam or being locked into
your e-mail communication system forever. Not only will posting this
information increase your opt-in rote, but if your company is
investigated for a spam complaint, then it may help show that you're a
permission-based company and not a spammer.
For the latest information about
spam
The Goals of spam
The goal of spam
is to determine the intrinsic grouping in a set of unlabeled data. But
how to decide what constitutes a good spam? It can be shown that
there is no absolute “best” criterion which would be independent of the
final aim of the spam. Consequently, it is the user which must
supply this criterion, in such a way that the result of the spam
will suit their needs.
For instance, we could be interested in finding representatives for
homogeneous groups (data reduction), in finding “natural
clusters” and describe their unknown properties (“natural” data
types), in finding useful and suitable groupings (“useful” data
classes) or in finding unusual data objects (outlier detection).
For the latest information about
The Goals of spam
Who uses spam?
Many different types of organizations use
spam as a vital
part of the work. A sampling of these include:
-
Marketing:
finding groups of customers with similar behavior given a large
database of customer data containing their properties and past
buying records;
-
Biology:
classification of plants and animals given their features;
-
Libraries:
book ordering;
-
Insurance:
identifying groups of motor insurance policy holders with a high
average claim cost; identifying frauds;
-
City-planning:
identifying groups of houses according to their house type,
value and geographical location;
-
Earthquake
studies: spam observed earthquake epicenters to
identify dangerous zones;
-
WWW:
document classification; spam weblog data to discover
groups of similar access patterns.
For the latest information about
Who Uses spam
Exchange with us