Sequence Alignment Algorithm in Similarity Measurement

The first and foremost question needed to be considered in clustering analysis is how to measure the similarity that decides the result of clustering immediately. However, are many shortcomings in traditional methods. This paper deals with similarity of English texts using sequence alignment which is always used in biology informatics. This method do not use traditional way that transform texts so that it is more intuitive. It can improve the rate and the result of clustering preferably. The test demonstrates the new approach is reasonable and efficient.