Showing papers on "Edit distance published in 1984"

The string-to-string correction problem with block moves

Abstract: The string-to-string correction problem is to find a minimal sequence of edit operations for changing a given string into another given string. Extant algorithms compute a longest common subsequence (LCS) of the two strings and then regard the characters not included in the LCS as the differences. However, an LCS does not necessarily include all possible matches, and therefore does not produce the shortest edit sequence. An algorithm that produces the shortest edit sequence transforming one string into another is presented. The algorithm is optimal in the sense that it generates a minimal covering set of common substrings of one string with respect to another. Two improvements of the basic algorithm are developed. The first improvement performs well on strings with few replicated symbols. The second improvement runs in time and space linear to the size of the input. Efficient algorithms for regenerating a string from an edit sequence are also presented.

Algorithms for String Editing which Permit Arbitrarily Complex Editing Constraints

Abstract: Let X and Y be any two strings of finite length. We consider the problem of transforming X to Y using the edit operations of deletion, insertion and substitution. The optimal transformation is the one which has the minimum edit distance associated with it. The problem of computing this distance and the optimal transformation using no edit constraints has been studied in the literature. In this paper we consider the problem of transform X to Y using any arbitrary edit constraints involving the number and type of edit operations to be performed. An algorithm has been presented to compute the minimum distance associated with editing X to Y subject to the specified constraint. The algorithm requires 0(|X|.|Y|.min(|X|, |Y|)) time and space. The technique to compute the optimal -transformation has also been presented.