Machine Learning with R

Example – filtering mobile phone spam with the naive Bayes algorithm

As worldwide use of mobile phones has grown, a new avenue for electronic junk mail has been opened for disreputable marketers. These advertisers utilize Short Message Service (SMS) text messages to target potential consumers with unwanted advertising known as SMS spam. This type of spam is particularly troublesome because, unlike email spam, many cellular phone users pay a fee per SMS received. Developing a classification algorithm that could filter SMS spam would provide a useful tool for cellular phone providers.

Since naive Bayes has been used successfully for email spam filtering, it seems likely that it could also be applied to SMS spam. However, relative to email spam, SMS ...