Spell-checking method is used to detect and correct the incorrect words which spelled in a document. To afford the best probable solution for corrected words first we should detect the incorrect words. The several spell correcting techniques are used after detecting the wrong words or misspelled words, to produces the best accurate correct or alternate words. Spell checking is used in word processing programs, cell phones, email programs and many other applications such as blogs and forums etc. Classification of Error correction and detection provide a better understanding of different types of techniques. There are two basic approaches to correct the spelling called edit distance approach and N-gram approach. This paper deals with a new technique called N-Gram based Edit distance (NGE) for transformation of strings. For rule generation to estimate the probabilities using Log model. The proposed method is compared with the other two approaches: Edit distance and N-Gram. The proposed method yields better result than other two approaches and the hybrid algorithm outperforms both edit distance and n gram algorithm in terms of efficiency and accuracy.