We describe a sorting algorithm that is optimally adaptive with respect to several important measures of presortedness. In particular, the algorithm requires O(n log(k/n)) time on sequences with k inversions; O(n-F k log k) time on sequences X that have a longest ascending subsequence of length n - k and for which Rera(X) = k; and O(nlog k) time on sequences that can be decomposed into k monotone shuffles. The algorithm makes use of an adaptive merging operation implemented using finger search trees.
CITATION STYLE
Moffat, A., Petersson, O., & Wormald, N. C. (1992). Sorting and/by merging finger trees. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 650 LNCS, pp. 499–508). Springer Verlag. https://doi.org/10.1007/3-540-56279-6_102
Mendeley helps you to discover research relevant for your work.