Lex2Emo

Installation

You can install the current version of Lex2Emo using devtools:

devtools::install_github("https://github.com/Reilly-ConceptsCognitionLab/Lex2Emo.git")

Reading in a directory of .txt files

Emo2Lex provides a method (readText) that allows users to concatonate a directory of .txt files into a single data frame meeting the format requirements for the next stage. Each file is read in as a single string of text, with only newlines removed, however non-alphanumeric characters are removed in the next stage.

Requirements for dataframe input

Data frame inputs must at the least contain a column with text data called ‘Text’ and an ID column (called ‘ID’) used to differentiate each text sample. Currently, an ID column is required even if a single text sample is provided. Lex2Emo will treat any data outside of the ‘Text’ and ‘ID’ columns as metadata and preserve the first entry for each document ID when aggregated.

Basic example of data frame for input into transformText()

ID	Text	Length
1	A description of something	55
2	The entirety of a book	80000
3	A whole conversation	700

Using transformText()

transformText() transforms each word in each text sample into it’s associated ratings (if available) on 22 emotional dimensions.

library(Lex2Emo)
# example of transformText() calls to output aggregated data
aggTextData <- transformText(textData = exampleData, returnFullData = F)

# example of transformText() calls to output non-aggregated data
aggTextData <- transformText(textData = exampleData, returnFullData = T)