Accuracy/speed trade-off technique for dynamic programing based algorithms

Abstract

Although the dynamic programming algorithm is an optimization approach used to solve a complex problem fast, the time required to solve it is still not efficient and grows exponentially as a function of the size of the input. In this contribution, we improve the computation time of the dynamic programing based algorithms by proposing new acceleration technique which splits the compared sequences into segments and then applies the dynamic programming algorithm to each segment individually. The new technique is adjustable for either the alignment accuracy or the computation speed. The results show that our technique performs much faster than the conventional dynamic programming algorithm with a slight degradation in the performance and provides much accurate results than the heuristic algorithm based on number of segments.