sample_topic_doc_probs: Per-document topic probabilities for the sample topics

In sdaume/topicsplorrr: A Package Supporting Topical Text Analysis, Exploration and Visualization

Description

This package data set summarizes the per document topic probabilities according to the stm topic model sample_topics fit to the inaugural presidential speeches in the respective dataset included in the quanteda package (quanteda::data_corpus_inaugural).

Usage

sample_topic_doc_probs

Format

a dataframe with 4 variables:

document

a unique document identifier, here a combination of president and year of speech

occur

the (assumed) date of document (i.e. speech) as a Date object

topic_id

a topic ID as assigned in the applied topic model

gamma

the probability that the given topic_id explains the given document; in the analysis we interpret this as the topic's share in this document

Details

This dataset is included for convenience and reproducibility, most analysis and visualization functions in this package rely on the per-document topic probabilities.

sdaume/topicsplorrr documentation built on Dec. 22, 2021, 11:11 p.m.