University’s Spam Detection System
Related Documents:
What is Spam?
Spam Quarantine Service
Tips & Tricks on reducing spam
E-mail: What are the hidden dangers?
What is phishing?
Information Technology (IT) has deployed a centralized spam detection system built into the campus' mail servers that scan for spam messages. The spam detection system looks at various parts of an e-mail message and will give it a percentage that it thinks is spam. Messages that are given a percentage of 100%, will be automatically rejected by the campus mail system before reaching a users mailbox and apply any rules/filters created by the user. As such, these messages will not be delivered to the users mailbox.
Users who have filters setup, including white/safe or block/deny listings will not be used for messages given a 100% rating by the spam detection system. Messages that do not have a 100% spam rating will be delivered to the users mailbox and follow any white or safe and block lists users have enabled or other filter rules.
The sender will receive a reject / undeliverable message or error in their e-mail client containing the following reason:
- 5.7.1 Command Rejected
Solution: The mail server thinks the content of the message is 100% spam. Resend the message to the recipient(s) and also add whitelist@rockefeller.edu in the To: field. E-mails sent to whitelist@rockefeller.edu are sent to the anti-spam vendor who will investigate the message and if deemed not to be spam, will update the spam signatures on the campus' mail server to allow future e-mails sent to/from a University e-mail address.
The recipient of the message will not be made aware that a message being addressed to their attention has been deleted by the campus mail system as being classified as 100% spam.
These messages are identified by, but not limited to, as spam:
- a "From" address that does not match the full header information of the message.
- a link that seemingly points to a well known Web site (such as Microsoft) but actually points to another address.
- keywords (such as "Buy Viagra") or other identifiers used by known spammers.
Messages that do not have a 100% spam rating will be delivered to the users mailbox and follow any white or safe and block lists users have enabled or other filter rules.
How Accurate is the University’s Spam Detection System?
Our anti-Spam technology accurately detects more than 90% of spam messages. The technology has even proved successful in deflecting e-mail-born spamming viruses such as the “Melissa” or “LoveBug” viruses. In evaluation testing conducted by Information Technology, accuracy rates have been as high as 99%. The false positive rate is a negligible .0001%, meaning fewer than 1 in 1 million legitimate e-mail messages are falsely identified as spam.
