A novel spell checking algorithm for non-segmented languages
Tim D. Hunt, Blaine Rakena, Kevin Wang · Wintec Research Archive (Waikato Institute of Technology) · 2013
This work describes for the first time a novel approach to implementing a successful spell checker for non-segmented languages such as Chinese. The design combines a novel technique that we call First Character Identifier Approach (FCIA). FCIA elegantly side steps the widely reported problem of how to segment text into words in a timely fashion. It does this by using a list of incorrectly spelt words to search for word matches in the text using the first character of the incorrect word. The design has been successfully implemented into a children’s email application that can be downloaded from http://www.mifrenz.com.