Choosing the best classifier for the job: Mobile Filtering for the South African Context

  • Marissa Griesel Centre for Text Technology (CTexT), Potchefstroom, South Africa
  • Wildrich Fourie Centre for Text Technology (CTexT), Potchefstroom, South Africa

Abstract

Short messages to cell phones (SMSs) have become the most popular means of communication on digital fronts, especially in Africa and South Africa in particular. This inspires the abuse of such systems by advertisers through the distribution of SPAM. It has therefore become necessary to incorporate a filtering system similar to e-mail classification on these low resource devices. In this article one practical method of filtering such messages for smart phones is described. A prototype application dubbed SpaMiNot intercepts incoming SMS messages and evaluates the message(s) with a number of filters. The application then classifies the message as legitimate (HAM) or unwanted (SPAM). HAM messages are left unhindered, while messages marked as SPAM can either be automatically deleted or stored in a SPAM list, from where it can be restored or deleted. The AndroidTM platform was chosen due to its recent and significant rise in popularity, as well as the ease and speed of Android application development. The application has been thoroughly tested on Android emulators and the VodafoneTM 845 device.

Published
2012-12-01
How to Cite
Griesel, M., & Fourie, W. (2012). Choosing the best classifier for the job: Mobile Filtering for the South African Context. Computational Linguistics in the Netherlands Journal, 2, 23-33. Retrieved from https://clinjournal.org/clinj/article/view/13
Section
Articles