unsakred.net

I am writing this post because of a rather heated debate going on over at the Yahoo! Messenger Blog.The general consensus of the comments is that Yahoo! is inundated with way to many SPAM messages and Bots.

This is entirely true and has been true for quite a long time now.

In response to the various posts complaining about Chat being flooded with porn bots and ads, I wrote a quick message about Bayesian spam filtering. For the most part this is the current method for catching E-mail SPAM. I don’t really see a huge problem with porting this over to the YMSG protocol. The problem being that you have to *train* the spam filter to detect what is really spam and what a valid message is.

Then it dawned on me. I think Yahoo! is already doing this, or if not they totally should be. Have you ever been to a chat room then get a few messages from the bots, and in the message there is a *this message is spam*? Now to me it would make sense that this is Yahoo!’s attempt at training the filter. It is an unobtrusive way to start learning valid messages / spam messages.

Now you might ask, why has Yahoo! not implemented it yet? Well there is well over a million users on the Yahoo! Messenger service, so one would think it would take some time to weed through the spam messages submitted, to make the filter at least some what operable. They could take it a step further and save some cpu time by creating it just for non friends. This way the chance of losing a message here and there won’t be nearly as high.

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

This entry was posted on Wednesday, March 14th, 2007 at 3:07 pm and is filed under Yahoo!, Yahoo! Messenger, Yahoo! Messenger, Yahoo! Services. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.