Efficient randomized pattern-matching algorithms：（高效的随机模式匹配算法）.pdfVIP

by Richard M. Karp Efficient Michael 0. Rabin randomized pattern-matching algorithms We present randomized algorithms to solve the We present a randomized algorithm to solve this problem. following string-matching problem and some of The algorithm associates with each stringX a fingerprint its generalizations: Given a string X of length n I#@) which is much shorter than the string itself. The search (the pattern)and a stringY (the text), find the for a match then compares short fingerprints insteadof long first occurrence of X as a consecutive block strings. within Y. The algorithms represent strings of The algorithm selects thejngerprintfunction at random length n by much shorter strings called from a family of easy-to-compute functions. No matter fingerprints, and achieve their efficiency by manipulating fingerprints instead of longer which input ((X(i),Y(i)))is presented, the algorithm is strings. The algorithms require a constant unlikely to produce afalse match, in which two fingerprints number of storage locations, and essentially run agree even though the original stringsdo not. in real time. They are conceptually simple and The most widely studied pattern-matching problem isthe easy to implement. The method readily following: Givenapattern Xof length n and a text Y of generalizes to higher-dimensional pattern- length m 2 n,find the first occurrence ofX as a consecutive matchin