【24h】

The Average Profile of Suffix Trees

机译:后缀树的平均剖面

获取原文

摘要

The internal profile of a tree structure denotes the number of internal nodes found at a specific level of the tree. Similarly, the external profile denotes the number of leaves on a level. The profile is of great interest because of its intimate connection to many other parameters of trees. For instance, the depth, fill-up level, height, path length, shortest path, and size of trees can each be interpreted in terms of the profile. The current study is motivated by the work of Park et al. [22], which was a comprehensive study of the profile of tries constructed from independent strings (also, each string generated by a memoryless source). In the present paper, however, we consider suffix trees, which are constructed from suffixes of a common string. The dependency between suffixes demands a careful, intricate treatment of overlaps in words. We precisely analyze the average internal and external profiles of suffix trees generated by a memoryless source. We utilize combinatorics on words (in particular, autocorrelation, i.e., the degree to which a word overlaps with itself) generating functions, singularity analysis, and the Mellin transform. We make comparisons of the average profile of suffix trees to the average profile of tries constructed from independent strings. We emphasize that our methods are extensible to higher moments. The present report describes the first moment of both the internal and external profiles of suffix trees.
机译:树结构的内部轮廓表示在树的特定级别上找到的内部节点的数量。类似地,外部轮廓表示水平上的叶子的数量。由于与树木的许多其他参数的密切相关,个人资料很有兴趣。例如,可以在轮廓方面解释树的深度,填充级别,高度,路径长度,路径长度,最短路径和大小。目前的研究是由Park等人的工作的推动。 [22],这是一种全面研究由独立字符串构建的尝试轮廓(也是由无记忆源生成的每个字符串)的概况。然而,在本文中,我们考虑后缀树,其由公共字符串的后缀构成。后缀之间的依赖性需要仔细,复杂的单词重叠处理。我们精确地分析了无记忆源生成的后缀树的平均内部和外部配置文件。我们利用单词(特别是自相关,即单词与本身重叠的程度)的组合学,产生功能,奇点分析和MELLIN变换。我们使后缀树的平均概况与独立字符串构造的尝试平均剖面进行比较。我们强调,我们的方法是更高时刻的可扩张性。本报告描述了后缀树的内部和外部轮廓的第一矩。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号