The Constrained Longest Common Subsequence: Theory and Experiments

Abstract

A common operation on strings is to calculate a value of similarity. The longest common subsequence (LCS) is a measure of similarity between two sequences. An extension of LCS measure is the constrained longest common subsequence (CLCS). The CLCS consists of calculating the LCS of two sequences, and it is a supersequence of a constrained sequence. This paper is an experimental evaluation of the most important CLCS algorithms. The implemented algorithms have a similar performance and were tested with different kinds of random strings belonging to the alphabets: generated uniformly at random (4, 10, 20, 32, 48 and 64 characters) and English text (239 characters). This study could help in selecting optimal algorithms for a given problem.