noise: detect noise

In nrauscher/corpus: Toolkit for Corpus Analysis

Description

detect noise

Usage

noise(.Object, ...)

## S4 method for signature 'DocumentTermMatrix'
noise(.Object, minTotal = 2,
  minTfIdfMean = 0.005, sparse = 0.995, stopwordsLanguage = "german",
  minNchar = 2, specialChars = getOption("polmineR.specialChars"),
  numbers = "^[0-9\\.,]+$", verbose = TRUE)

## S4 method for signature 'TermDocumentMatrix'
noise(.Object, ...)

## S4 method for signature 'character'
noise(.Object, stopwordsLanguage = "german",
  minNchar = 2, specialChars = getOption("polmineR.specialChars"),
  numbers = "^[0-9\\.,]+$", verbose = TRUE)

## S4 method for signature 'textstat'
noise(.Object, pAttribute, ...)

Arguments

.Object	an .Object of class "DocumentTermMatrix"
...	further parameters
minTotal	minimum colsum (for DocumentTermMatrix) to qualify a term as non-noise
minTfIdfMean	minimum mean value for tf-idf to qualify a term as non-noise
sparse	will be passed into "removeSparseTerms" from "tm"-package
stopwordsLanguage	e.g. "german", to get stopwords defined in the tm package
minNchar	min char length ti qualify a term as non-noise
specialChars	special characters to drop
numbers	regex, to drop numbers
verbose	logical
pAttribute	relevant if applied to a textstat object

Value

a list

nrauscher/corpus documentation built on May 23, 2019, 9:34 p.m.