Stop words for “Dr Math”

摘要

“Dr Math” is a facility where primary and secondary school pupils can use MXit on their cell phones to get help with their mathematics homework. Pupils use an abbreviated “MXit lingo” leaving out most vowels and substituting various numerals and symbols for common sounds. Topic spotting in these conversations is required for two reasons. From a pedagogical point of view, it would be beneficial to determine if high numbers of pupils were needing help with similar topics. This could indicate a lack or gap in the educational materials. Another important use would be to attempt to pick up any inappropriate conversations between pupil and tutor which may not adhere to the code of conduct signed by the tutors or to the ethical clearance of the original “Dr Math” project. This paper describes work in spotting topics in conversations between pupils and tutors. This work is to first determine which words can be safely ignored by the future topic spotters. These words which can be ignored are known as “stop words”.