text_to_orgs: Match messy text data to social organizations using a...

In brandonleekramer/tidyorgs: Standardize messy text data for organizational analysis

Description

This function matches unstructured text data to various dictionaries of organizations by extracting and iterating through consecuetive word sequences (or n-grams). To do this, the function extracts n-grams using the tidytext package, matching all sequences in the unstructured data that have n words and then 'funneling' through all sequences of n-1, n-2, etc. words before matching the single tokens. This process returns a dataframe of ids, organizations, and sectors for only those rows matched within the sectors specified.

Usage

text_to_orgs(
  data,
  id,
  input,
  output,
  sector = c("academic", "business", "government", "nonprofit")
)

Arguments

data	A data frame or data frame extension (e.g. a tibble).
id	A numeric or character vector unique to each entry.
input	Character vector of messy or unstructured text that will be unnested as n-grams and matched to dictionary of organizations in specified sector.
output	Output column to be created as string or symbol.
sector	Sector to match by organizations. Currently, the only option is "academic" with "business", "government", "household", and "nonprofit" in development.

Examples

library(tidyverse)
library(tidyorgs)
data(github_users)

classified_by_text <- github_users %>%
  text_to_orgs(login, company, organization, academic)

brandonleekramer/tidyorgs documentation built on Dec. 19, 2021, 11:42 a.m.