computeFrequencies: Normalized Frequencies.

In keblu/GWP: Generalized Word Power Approach

Description

Generate the Normalized Frequency table from textual data input.

Usage

computeFrequencies(corpus, sentimentWord, shifterWord, clusterSize = 1)

Arguments

corpus	corpus with column docID, regID, and texts. docID indicate a unique ID for the document. regID is used for aggregation of document for the regression. It must match with the specific regID of the response variable in fitGWP. texts are the textual data.
sentimentWord	Vector of words used for computing the sentiment.
shifterWord	Matrix with element x and y. x is a vector of valence shifting words while y is the modifier values.
clusterSize	Scalar indicating the window in which valance shifting words have an influence.

Value

A list with the following elements:

• docID: unique document ID.

• regID: regression ID.

• loc: location of the sentiment word within the texts.

• word: sentiment word.

• shift: shigt coefficient modifying the sentiment word.

• NormalizedFrequency: frequency of word normalized by number of token in each text

• NormalizedFrequencyPerRegID: Normalzied Frequency normalized by the number of documents per each regression ID.

Examples

# Load example data
data("corpus",  package = "GWP")

# Setup the lexicons
sentimentWord <- sentometrics::list_lexicons$LM_en$x
shifterWord <- sentometrics::list_valence_shifters$en[, c("x", "y")]

# Generate the frequency data
frequencies <- computeFrequencies(corpus, sentimentWord, shifterWord, clusterSize = 5)

keblu/GWP documentation built on June 15, 2020, 11:43 p.m.