JSTOR_lda_hotncoldtopics: Plot the top five hot and cold topics in the archive

In benmarwick/JSTORr: Simple Text Mining and Document Clustering of JSTOR Journal Articles

Description

Generates plots and data frames of the top five hot and cold topics. Hot topics are topics with a positive correlation to year of publication, cold topics have a negative correlation. For use with JSTOR's Data for Research datasets (http://dfr.jstor.org/).

Usage

JSTOR_lda_hotncoldtopics(lda, pval = 0.05, ma = 5, size = 12)

Arguments

lda	the object returned by the function JSTOR_lda.
pval	p-value of the correlation cutoff for topics to include in the top 5 negative/positive list (ie. only topics where p<0.01 or 0.001?). Default is 0.05.
ma	moving average interval, default is five years.
size	size of the text in the legend (the topic keywords). Default is 12.

Value

Returns a plot of the hot topics and plot of the cold topics and a list of dataframes of the topic proportions per year. Years as rows, topics as columns and posterior probabilities as cell values. Uses a five year moving average to smooth the plots a bit.

Examples

## hotncold <- JSTOR_lda_hotncoldtopics(lda = lda150, ma = 10) 

benmarwick/JSTORr documentation built on May 12, 2019, 12:59 p.m.