A two-phase method of approximate string match

Approximate string match (ASM) is the key technique in text correction and some kinds of information retrieval application. The major problem in ASM is the efficiency of the process to find the similar words in a large vocabulary. This paper proposed a two phase method to improve the entire efficiency of ASM. A given string can be located at a point in a multidimensional word space that is organized by the feature vector of words, and then its neighbors are to be compared with the given string more accurately to get the final candidate list. Experiments have been conducted and the results show that the proposed algorithm can effectively improve the entire efficiency of approximate string match.