基于后缀数组的字符串模式查找的算法-计算机软件与理论专业论文.docxVIP

III III Abstract Pattern-matching problem is a fundamental problem in computer science. In the early years, most of pattern-matching algorithems focus on the exact pattern-matching algorithms study, such as: well-known single pattern matching algorithms KMP BM and multi-pattern matching algorithms CA CW BNDM so. However, in practical applications, it is very important that approximate pattern matching in information filtering and computational biology and other fields. For example, in the analysis of DAN fragments, people usually search similar patterns with the target DNA fragment, in order to determine whether there are genetic relations between them. To search a meaningful pattern string quickly in the vast amounts of information, we must improve the efficiency of pattern matching algorithms. Therefore, it is necessary to improve pattern-matching algorithms furtherly. Here for the problem of searching the high-frequency words (pattern string) in longer English novel, we applied filtering ideas to present approximate string pattern searching algorithms Epattern_searcherH and Epattern_searcher. The main work of this paper is as follows: Propose the algorithms of pattern searching, for the idea of the filter design pattern searching algorithms of approximate pattern matching. First, according to certain filter principles (limited to certain conditions) to search a text string, filter out those unlikely matches the text segment; and then further verify that the positions of the rest of the candidate matches whether it really exists at the success of matching. As filtered out most of unlikely match the text string, the process of matching has greatly accelerated. To achieve the pattern searching algorithms based on suffix arrays data structure.Suffix tree is a good data structure for the treatment of large-scale string, but the requiems of suffix tree index structure space is too large, thus there is a pattern searching index structure—— suffix arrays. Because the sto