Algoritmy vyhledávání v textu

Abstract

The aim of this bachelor thesis is implementation of library for approximate pattern matching. This library will allow seeking of user specified pattern with specified number of maximum mistakes in text based on deterministic and nondeterministic finite automata. User will be able to choose between Hamming distance and Levenshtein distance. The first part describes use of finite automata for approximate pattern matching. The second part describes implementation of libraries. The third part focuses on experiments with implemented libraries. The conclusion of this thesis summarizes advantages and disadvantages of this approach to approximate pattern matching.