filter_dfm: Filter a document-feature matrix for systematic language...

In chainsawriot/rectr: Reproducible Extraction of Cross-lingual Topics Using R

Filter a document-feature matrix for systematic language differences

Description

This function filters a document-feature matrix using singular value decomposition.

Usage

filter_dfm(
  input_dfm,
  k,
  corpus = NULL,
  multiplication_factor = 2,
  dimension = 100,
  alpha = 0.05,
  noise = FALSE
)

Arguments

input_dfm	dfm generated by dfm_boe()
k	integer, number of topics
corpus	a multilingual corpus generated by create_corpus()
multiplication_factor	integer, select k * mulitiplication_factor columns from the U-Matrix.
dimension	integer, the first singular value to be extracted in the U-Matrix.
alpha	double, alpha level to filter the U-Matrix using one-way ANOVA.

Value

an rectr_dfm object

chainsawriot/rectr documentation built on July 30, 2023, 2:30 p.m.