cor.network: Network graph of the most correlated pairs of terms found...

In JSmith146/CoRpEx: CorpEx is Designed for the Exploration and Mining of Large Textual Corpuses of Documents

Description

This function visualizes the most frequent pair of terms found throughout a corpus of documents based on the Phi Correlation Coefficient. The function will only produce word pairs at or above a provided correlation limit.

Usage

cor.network(data.td, corlimit)

Arguments

data.td	A tidy dataset
corlimit	A user defined numeric correlation limit between 0 and 1

Details

This function identifies all of the terms found with in a corpus and sums the number of times each term is found within the same document relative to the other terms. With this information, each pair of terms is provided a correlation value using the Phi Coefficient. The function then generates an undirected network graph connecting each word (node) in a word pair. Correlated words are then connected by edges (lines) whose thickness varies based on the value of the correlation coefficient for each word pair.

Value

A network graph of the most correlated word pairs down to a user defined corlimit

Examples

cor.network(Articles, 200)

JSmith146/CoRpEx documentation built on May 17, 2019, 10:11 p.m.