Similarity Joins in Relational Database Systems

Since the edit distance is computationally expensive, token-based distances have been introduced to speed up edit distance computations. Token-based distances are used to compute an approximation of the edit distance and prune expensive edit distance calculations.