This workflow reads in a series of misspelled retailer names, along with the desired version of the names. Using the Similarity Search node, the closest correctly spelled name is matched based on Levenshtein distance.