Chapter 7. Lupus. Textual Data Science With R

1. script.lupus

script.lupus.R in UTF-8 format can be downloaded from:

File lupus-506.csv in UTF-8 format can be downloaded from the Internet and save it to some directory:

2. Loading Xplortext package

3. Loading the database

4. PART I

Exploratoria analysis

Correspondence analysis

Barchart of the first ten eigenvalues

Metakeys and Dockeys

Plot of planes (1,2) and (3,4)

PART II

Analysis years x words

Display of a part of the aggregate table

Initial statistics

Display of the number of documents of every year

Correspondence analysis of the ALT years*words

Barchart of the eigenvalues

Representation of the years

Representation of the words

PART III

Analysis of the table years x drug names

setA: all the words, excepted the stopwords

setB: drugnames cited in the database

Nomedic: non-drugs which have to be eliminated

Building the table years x drugs

Characteristic drugs for the year