Skip to main content

Table 1 Column descriptions of the custom dictionary

From: A UMLS-based spell checker for natural language processing in vaccine safety

Column

Description

word_id

Unique identifier

word_str

Dictionary word

word_ngram

Bigrams of the dictionary word. Example: "pediatrician" would have the following bigrams: pe, ed, di, ia, at, tr, ri, ic, ci, ia, an

word_metaphone

The metaphone value of the dictionary word. Example: pediatrician would have the metaphone PTTRXN

word_header

The first 4 characters of the word. Example: "pediatrician" would have the header pedi

word_anterior

The 4 characters after the first character of the dictionary word

word_posterior

The 4 characters before the last character of the dictionary word

word_fragment

If the dictionary word is longer than 10 characters the first 10 characters of the dictionary word