M4ttoF / MED-Spell-Correction

Given a dictionary D and a spelling error corpus C for the English language, calculates the average success at k (s@k) for the minimum edit distance (MED) algorithm for all misspelled tokens in C

MED-Spell-Correction

Takes a list of incorrect spellings of words from Birkbeck corpus and finds Levenshtein distance with words from WordNet dictionary

Requirements

If you are using conda you can use the environment.yml file Otherwise all modules used are listed in requirements.txt

Running program

After installing the dependencies run as python parallel.py It will give you the S@K score for K=1,5,10 accordingly