Indexed by:
Abstract:
In this work, we consider the matrix chain ordering problem to determine the optimal computation order of the matrix chain products. A new algorithm for the matrix chain ordering problem is presented. The time complexity of the presented algorithm is O(n logm), where n is the number of matrices in the chain and m is the number of local minimums in the dimension sequence of the given matrix chain. When m is a fixed constant, the new algorithm requires only O(n) time. The new algorithm is not only an improvement on the time complexity compared to the O(n log n) time algorithm of Hu and Shing, but also substantially simpler than the Hu-Shing algorithm. 1553-9105/Copyright © 2014 Binary Information Press.
Keyword:
Reprint 's Address:
Email:
Version:
Source :
Journal of Computational Information Systems
ISSN: 1553-9105
Year: 2014
Issue: 10
Volume: 10
Page: 4299-4306
Cited Count:
SCOPUS Cited Count: 1
ESI Highly Cited Papers on the List: 0 Unfold All
WanFang Cited Count:
Chinese Cited Count:
30 Days PV: 2
Affiliated Colleges: