%0 Journal Article
%T Using Modified Fixed Point Numbers Representations in Spelling Checker
%A R.F. Alwan
%A O. Al-Hadithi
%J Journal of Applied Sciences
%D 2010
%I Asian Network for Scientific Information
%X This study is concerned with the process of designing and implementing an English text spelling checker which can detect and correct the misspelled words. The spelling checker is built using a modified version of Fixed Point Numbers Representations Technique (FPNRT), which is a compression method that is suitable to this kind of applications. The modified FPNRT will transform the dictionary words into numeric values and then stores them in a data base file of that form, which is considered to be a new way for storing a dictionary. This method gives comparatively good compression ratio and very quick in compression/decompression process. Therefore, this method is good for files which are frequently read and updated. This approach will minimize the memory space required to store the dictionary by achieving a compression ratio of 37% on average and also enhance the performance of the spelling checker by maximizing its speed which takes O(lg n/5). This approach is also provide a new way for building a list of candidate words that may be used to correct the misspelled words.
%K dictionary based
%K statistical-analysis methods
%K spelling errors
%K data compression
%K spelling corrector
%K Spelling checker
%U http://docsdrive.com/pdfs/ansinet/jas/2010/1963-1967.pdf