An efficient algorithm for computing the edit distance of a regular language via input-altering transducers

We revisit the problem of computing the edit distance of a regular language given via an NFA. This problem relates to the inherent maximal error-detecting capability of the language in question. We present an efficient algorithm for solving this problem which executes in time $O(r^2n^2d)$, where $r$ is the cardinality of the alphabet involved, $n$ is the number of transitions in the given NFA, and $d$ is the computed edit distance. We have implemented the algorithm and present here performance tests. The correctness of the algorithm is based on the result (also presented here) that the particular error-detection property related to our problem can be defined via an input-altering transducer.