Package 'vader'

Title: Valence Aware Dictionary and sEntiment Reasoner (VADER)
Description: A lexicon and rule-based sentiment analysis tool that is specifically attuned to sentiments expressed in social media, and works well on texts from other domains. Hutto & Gilbert (2014) <https://www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/view/8109/8122>.
Authors: Katherine Roehrick [aut, cre]
Maintainer: Katherine Roehrick <[email protected]>
License: MIT + file LICENSE
Version: 0.2.1
Built: 2024-11-13 05:14:27 UTC
Source: https://github.com/cran/vader

Help Index

Get a named vector of vader results for a single text document

Description

Use get_vader() to calculate the valence of a single text document.

Usage

get_vader(text, incl_nt = T, neu_set = T, rm_qm = T)

Arguments

text

to be analyzed; for get_vader(), the text should be a character string
incl_nt

defaults to T, indicates whether you wish to incl UNUSUAL n't contractions (e.g., yesn't) in negation analysis
neu_set

defaults to T, indicates whether you wish to count neutral words in calculations
rm_qm

defaults to T, indicates whether you wish to clean quotation marks from text (setting to F may result in errors)

Value

A named vector containing the valence score for each word; an overall, compound valence score for the text; the weighted percentage of positive, negative, and neutral words in the text; and the frequency of the word "but".

References

For the original Python Code, please see:

  • https://github.com/cjhutto/vaderSentiment

  • https://github.com/cjhutto/vaderSentiment/blob/master/vaderSentiment/vaderSentiment.py

For the original R Code, please see:

  • https://github.com/nrguimaraes/sentimentSetsR/blob/master/R/ruleBasedSentimentFunctions.R

Modifications to the above scripts include, but are not limited to:

  • ALL CAPS fx: updated to account for non-alpha words; i.e. "I'M 100 PERCENT SURE" would previously have been counted as mixed case due to the use of numbers

  • IDIOMS fx: added capacity to check for idioms that do not contain any words found in the Vader Lexicon

  • WORDS+EMOT: strip punctuation while preserving ALL emoticons found in dictionary

  • Option to turn on/off neutral count

N.B.

In the examples below, "yesn't" is an internet neologism meaning "no", "maybe yes, maybe no", "didn't", etc.

See Also

vader_df to get vader results for multiple text documents

Examples

get_vader("I yesn't like it")
get_vader("I yesn't like it", incl_nt = FALSE)
get_vader("I yesn't like it", neu_set = FALSE)
get_vader("I said \"I'm not happy\"", rm_qm = FALSE)
get_vader("I said \" I'm not happy \" ", rm_qm = FALSE)

Get a dataframe of vader results for multiple text documents

Description

Use vader_df() to calculate the valence of multiple texts contained within a vector or column in a dataframe.

Usage

vader_df(text, incl_nt = T, neu_set = T, rm_qm = F)

Arguments

text

to be analyzed; for vader_df(), the text should be a single vector (e.g. 1 column)
incl_nt

defaults to T, indicates whether you wish to incl UNUSUAL n't contractions (e.g., yesn't) in negation analysis
neu_set

defaults to T, indicates whether you wish to count neutral words in calculations
rm_qm

defaults to T, indicates whether you wish to clean quotation marks from text (setting to F may result in errors)

Value

A dataframe containing the valence score for each word; an overall, compound valence score for the text; the weighted percentage of positive, negative, and neutral words in the text; and the frequency of the word "but".

N.B.

In the examples below, "yesn't" is an internet neologism meaning "no", "maybe yes, maybe no", "didn't", etc.

See Also

get_vader to get vader results for a single text document