String matching with mismatches is a basic concept of information retrieval with some kinds of approximation. This paper proposes an FFT-based algorithm for the problem of string matching with mismatches, which computes an estimate with accuracy. The algorithm consists of FFT computations for binary vectors which can be computed faster than the computation for vectors of complex numbers. Therefore, a reduction of the computation time is obtained by the speed-up for FFT, which leads an improvement of the variance of the estimates. This paper analyzes the variance of the estimates in the algorithm and compares it with the variances in existing algorithms. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Baba, K. (2010). String matching with mismatches by real-valued FFT. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6019 LNCS, pp. 273–283). Springer Verlag. https://doi.org/10.1007/978-3-642-12189-0_24
Mendeley helps you to discover research relevant for your work.