syuzhet (version 1.0.7)

mixed_messages: Mixed Messages

Description

This function calculates the "emotional entropy" of a string based on the amount of conflicting valence. Emotional entropy is a measure of unpredictability and surprise based on the consistency or inconsistency of the emotional language in a given string. A string with conflicting emotional language may be said to express a "mixed message."

Usage

mixed_messages(string, remove_neutral = TRUE)

Value

A vector containing two named values

Arguments

string

A string of words

remove_neutral

Logical indicating whether or not to remove words with neutral valence before computing the emotional entropy of the string. Default is TRUE

Examples

text_v <- "That's the love and the hate of it" 
mixed_messages(text_v) # [1] 1.0 0.5 = high (1.0, 0.5) entropy
mixed_messages(text_v, TRUE)
# Example of a predictable message i.e. no surprise
text_v <- "I absolutley love, love, love it." 
mixed_messages(text_v) # [1] 0 0 = low entropy e.g. totally consistent emotion, i.e. no surprise
mixed_messages(text_v, FALSE)
# A more realistic example with a lot of mixed emotion.
text_v <- "I loved the way he looked at me but I hated that he was no longer my lover"
mixed_messages(text_v) # [1] 0.91829583 0.05101644 pretty high entropy.
mixed_messages(text_v, FALSE)
# A more realistic example without a lot of mixed emotion.
text_v <- "I loved the way he looked at me and I was happy that he was my lover."
mixed_messages(text_v) # [1] 0 0 low entropy, no surprise.
mixed_messages(text_v, FALSE)
# An urealistic example with a lot of mixed emotion.
text_v <- "I loved, hated and despised the way he looked at me and 
I was happy as hell that he was my white hot lover."
mixed_messages(text_v)
mixed_messages(text_v, FALSE)