spelling-corrector

a simple spelling corrector in python.

Based on the work of Peter Norvig.

Steps:

run generate_word_count.py to create the WORDS pickle file.
run unit_tests.py to make sure all the tests pass.
run spell_test.py on the given test files and get the accuracy of the corrector.

Note that the datasets directory contains the required data in .txt formats:

big.txt is a concatenation of public domain book excerpts from Project Gutenberg and lists of most frequent words from Wiktionary and the British National Corpus.
spell-testset1.txt and spell-testset2.txt are test files containing both correctly spelled words and their incorrect variations extracted from Birkbeck spelling error corpus

More info on the algorithm.