Here are the most commonly used algorithms and auxiliary utilities for fuzzy (similarity) string search in large dictionaries written in Java.

• Levenshtein Distance (with cutoff and prefix version)
• Damerau-Levenshtein Distance (with cutoff and prefix version)
• Extension (Spell-checker) Method
• N-Gram Method (with some modifications)
• Signature Hash Method
• Bitap (Shift-Or with Wu-Manber modifications)
• Burkhard-Keller (BK) Trees
• Skip algorithm

All implementations are aimed to provide simplicity and clarity of algorithm's work.

Related articles are at http://ntz-develop.blogspot.com/