The MEDLINE N-gram Set 2018: by Split, Group, Filter, and Combine Algorithm
The MEDLINE n-gram set (generated by split, group, filter, and combine algorithm) is listed as bellows. For each MEDLINE record, title and abstract are used as the source of n-grams. They are combined, tokenized into sentences, and then tokenized into tokens (words use space as word boundary). Finally, n-grams are generated by filtering out terms with more than 50 characters or the total word count is less than 30. The specifications of generating these n-grams are listed as follows:
Document count | Word count | N-gram |
N-grams | File | Zip Size | Actual Size | No. of n-grams |
---|---|---|---|---|
Unigrams | 1-gram.2018.tgz | 6.9 MB | 17 MB | 1,022,412 |
Bigrams | 2-gram.2018.tgz | 44 MB | 128 MB | 6,000,100 |
Trigrams | 3-gram.2018.tgz | 66 MB | 219 MB | 8,534,524 |
Four-grams | 4-gram.2018.tgz | 44 MB | 162 MB | 5,348,132 |
Five-grams | 5-gram.2018.tgz | 21 MB | 80 MB | 2,265,965 |
N-gram Set | nGramSet.2018.30.tgz | 184 MB | 603 MB | 23,171,133 |
Distilled N-gram Set | distilledNGram.2018.tgz | 72 MB | 234 MB | 8,979,895 |