...
首页> 外文期刊>Journal of the Association for Computing Machinery >Improved Bounds on the Average Length of Longest Common Subsequences
【24h】

Improved Bounds on the Average Length of Longest Common Subsequences

机译:最长公共子序列平均长度的改进界限

获取原文
获取原文并翻译 | 示例
           

摘要

It has long been known [Chvatal and Sankoff 1975] that the average length of the longest common subsequence of two random strings of length n over an alphabet of size k is asymptotic to yγ_kn for some constant γ_k depending on k. The value of these constants remains unknown, and a number of papers have proved upper and lower bounds on them. We discuss techniques, involving numerical calculations with recurrences on many variables, for determining lower and upper bounds on these constants. To our knowledge, the previous best-known lower and upper bounds for γ_2 were those of Dancfk and Paterson, approximately 0.773911 and 0.837623 [Dancfk 1994; Dancik and Paterson 1995]. We improve these to 0.788071 and 0.826280. This upper bound is less than the γ_2 given by Steele's old conjecture (see Steele [1997, page 3]) that γ_2 = 2/(1 + 2~(1/2)) ≈ 0.828427. (As Steele points out, experimental evidence had already suggested that this conjectured value was too high.) Finally, we show that the upper bound technique described here could be used to produce, for any k, a sequence of upper bounds converging to γ_k, though the computation time grows very quickly as better bounds are guaranteed.
机译:[Chvatal and Sankoff 1975]早就知道,在大小为k的字母上,长度为n的两个随机串的最长共同子序列的平均长度对于取决于k的某个常数γ_k渐近于yγ_kn。这些常数的值仍然未知,许多论文已经证明了它们的上下边界。我们讨论了涉及对许多变量进行递归的数值计算的技术,以确定这些常数的上下限。据我们所知,以前最著名的γ_2的上下界是Dancfk和Paterson的上下界,大约为0.773911和0.837623 [Dancfk 1994; 1991年。 Dancik and Paterson 1995]。我们将其提高到0.788071和0.826280。该上限小于斯蒂尔的旧猜想所给出的γ_2(参见斯蒂尔[1997,第3页]),即γ_2= 2 /(1 + 2〜(1/2))≈0.828427。 (正如斯蒂尔指出的那样,实验证据已经表明该推测值太高。)最后,我们证明了对于任何k,可以使用此处描述的上限技术来生成收敛到γ_k的一系列上限,尽管由于保证了更好的边界,所以计算时间增长很快。

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