Mobile Spam short analysis

Mobile Spam, that is, unwanted SMS or email messages received at the cell phone, is an increasing trend. A recent news item at the New York Times discusses the problem of Mobile Spam, with a sample of statistics, damages, and actions.

Statistics

According to Ferris Research, phone users will be receiving around 1.5 billion spam messages at their cell phones on 2008. If 48 billion messages were sent on December 2007, the rule is that 0.2% of the total messages will be spam. Just compare this number with the more than 80% of current email spam rates.

Sprint is currently blocking more than 65% of all text messages as spam, at the carrier itself.

An attack from a spammer who sent billions of messages through msn.com to phone email addresses at Verizon got more than 5 billion messages on March 2008 before the attack was stopped and blocked. This was done with a dictionary attack to all phone numbers followed with the provider standard @domain name.

Damages

For the end user:

• Cost - you may pay up to 20 cts. per spam received.
• Annoyment - you may receive the messages when working, or even when sleeping.
• Time - you loose your time deleting them.
• Miss legitimate messages - you get used with deleting messages, and you may delete some legitimate, important one.

For the carrier/service provider:

• Infrastructure and other resources- you have to increase your processing power to cope with the overhead, pay for an anti-spam solution, increase your administration manpower.
• Legal - you may get demanded for refunds by end users.
• Branding - you may get brand damage and loose customers.

Actions

The measures taken by carriers at the US are:

• Share information among providers - collaboration.
• Deploy spam filters at the network.
• Providing end users with configuration tools, which enable them to block all access to Internet, or to have a less obvious email accounts.

These measures are scarce (as the problem has not yet revealed as critical). But at Japan, the problem of email spam sent to mobile phones fully emerged some years ago, and a combination of technical measures, government legislation, and providers' self-regulation made it under control. See this presentation by Toshihiko Shibuya at the ITU WSIS Thematic Meeting on Countering Spam (July 2004).

Notes

I was been involved in a research conducted for Vodafone (2006-07), focused on Bayesian filtering for SMS spam. The main findings are in these papers:

Cormack, G. V., Gómez Hidalgo, J. M., and Puertas Sanz, E. 2007. Spam filtering for short messages. In Proceedings of the Sixteenth ACM Conference on Conference on information and Knowledge Management (Lisbon, Portugal, November 06 - 10, 2007). CIKM '07. ACM, New York, NY, 313-320. DOI= http://doi.acm.org/10.1145/1321440.1321486.

Cormack, G. V., Gómez Hidalgo, J. M., and Puertas Sanz, E. 2007. Feature engineering for mobile (SMS) spam filtering. In Proceedings of the 30th Annual international ACM SIGIR Conference on Research and Development in information Retrieval (Amsterdam, The Netherlands, July 23 - 27, 2007). SIGIR '07. ACM, New York, NY, 871-872. DOI= http://doi.acm.org/10.1145/1277741.1277951

Gómez Hidalgo, J.M., Cajigas Bringas, G., Puertas Sanz, E., Carrero García, F. Content Based SMS Spam Filtering. Dick Bulterman, David F. Brailsford (Eds.), Proceedings of the 2006 ACM Symposium on Document Engineering, Amsterdam, The Netherlands, ACM Press. Ámsterdam, The Netherlands, October 10-13, 2006.