Flexible Pattern Matching in Strings

Abstract

String matching problems range from the relatively simple task of searching a single text for a string of characters to searching a database for approximate occurrences of a complex pattern. Recent years have witnessed a dramatic increase of interest in sophisticated string matching problems, especially in information retrieval and computational biology. This book presents a practical approach to string matching problems, focusing on the algorithms and implementations that perform best in practice. It covers searching for simple, multiple and extended strings, as well as regular expressions, and exact and approximate searching. It includes all the most significant new developments in complex pattern searching. The clear explanations, step-by-step examples, algorithm pseudocode, and implementation efficiency maps will enable researchers, professionals and students in bioinformatics, computer science, and software engineering to choose the most appropriate algorithms for their applications.

A Dynamically Reconfigurable FPGA-Based Pattern Matching Hardware for Subclasses of Regular Expressions

“…The key to our architecture is the use of bit-parallel pattern matching approach developed in string matching communities since 1990s [2], [16], [23]. In Fig.…”

“…1, we show the top-level view of our architecture, which will be explained in detail. Our target pattern class is the class EXT of extended patterns [16], [17], which is a subclass of regular expressions allowing union of characters and its Kleene-star. The class EXT is a natural generalization of the class STR of exact string patterns consisting characters and concatenation, which supports character classes β = [ab · · · ], gaps '.…”

“…• Dynamic loading of extended patterns: Our Dynamic BP-NFA is the first dynamically reconfigurable hardware that can deal with the class EXT of extended patterns, which is a non-trivial and useful subclass of regular expressions widely studied in real-world applications such as event stream processing [1] and bioinformation processing [16], [17]. [3] is only a dynamically reconfigurable hardware before ours that can deal with a nontrivial subclass of regular expressions.…”

“…In the static compilation approach [19], [20], [24]- [26], a set of input regular expressions are transformed into either deterministic finite automata (DFAs) or non-deterministic finite automata (NFAs) [16], and then statically compiled into wired logic on FPGA. However, the static compilation approach has a drawback in that modification of regular expressions is too expensive to be done frequently.…”

“…, and Kleenestar R = (R 1 ) * , where R 1 and R 2 are regular expressions [16]. For a regular expression R, we denote by…”