manifestoR (version 1.5.0)

ManifestoDocument: Manifesto Document

Description

A ManifestoDocument represents a document from the Manifesto Corpus and contains text, coding and meta information. ManifestoDocument objects need not be constructed manually but are the content of the ManifestoCorpus objects downloaded from the Manifesto Corpus Database API via mp_corpus.

ManifestoDocuments subclass the TextDocument class from the package tm. Hence they can be and usually are collected in a tm Corpus to interface easily with text mining and other linguistic analysis functions. manifestoR uses the subclass ManifestoCorpus of tms Corpus, but ManifestoDocuments can be stored in any kind of Corpus.

As in tm any ManifestoDocument has metadata which can be accessed and modified via the meta function, as well as content, accessible via content. Additionally, via codes(), the coding of the (quasi-)sentence ccording to the CMP category scheme can be accessed (and modified). The CMP category scheme can be found online at https://manifesto-project.wzb.eu/coding_schemes/mp_v4 (version 4) or https://manifesto-project.wzb.eu/coding_schemes/mp_v5 (version 5).

Usage

ManifestoDocument(
  content = data.frame(),
  id = character(0),
  meta = ManifestoDocumentMeta()
)

Arguments

content

data.frame of text and codes for the ManifestoDocument to be constructed. There can be multiple columns of codes, but by default the accessor method codes searches for the column named "cmp_code".

id

an id to identify the Document

meta

an object of class ManifestoDocumentMeta containing the metadata for this document

Details

Internally, a ManifestoDocument is a data.frame with a row for every quasi-sentence and the columns text and code.

Examples

# NOT RUN {
corpus <- mp_corpus(subset(mp_maindataset(), countryname == "New Zealand"))
doc <- corpus[[1]]
print(doc)
# }