PolMine

library(polmineR)
use("GermaParl")

stopwords <- unname(unlist(
  noise(
    terms("GERMAPARL", p_attribute = "word"),
    stopwordsLanguage = "en"
  )
))

PolMine

Sys.setenv(
  "AWS_ACCESS_KEY_ID" = "<my-access-key>",
  "AWS_SECRET_ACCESS_KEY" = "<my-secret-access-key>"
)

library(aws.s3)
get_bucket("polmine")
lda <- s3readRDS("corpora/cwb/germaparl/germaparl_lda_speeches_250.rds", bucket = "polmine")

PolMine

# This code, which can be adapted easily, can be used to train a word2vec model easily. Note that it
# relies on the package [wordVectors](https://github.com/bmschmidt/wordVectors).


library(wordVectors)

file_out <- "~/Lab/tmp/germaparl.txt"
vectors_bin <- "~/Lab/tmp/germaparl.bin"

.fn <- function(x){

PolMine

library(magick)
library(purrr)

list.files(path = "~/Lab/tmp/", pattern = "*.png", full.names = T) %>% 
  map(image_read) %>% # reads each path file
  image_join() %>% # joins image
  image_animate(fps = 1) %>% # animates, can opt for number of loops
  image_write("~/Lab/annotation_demo.gif") # write to current dir

PolMine

# install current version of cwbtools

library(drat)
drat::addRepo("polmine")
install.packages("cwbtools")


# Reencode installed corpus

library(polmineR)

PolMine

library(cooccurrences)
library(pbapply)
library(coop)


issues <- df %>% unlist() %>% unname() %>% as.character() %>% unique()
dt <- count("GERMAPARL", issues) %>%
  setkeyv("count") %>% setorderv(cols = "count", order = -1L)
issues_min <- dt[count > 100][["query"]]
issues_min <- iconv(issues_min, from = "latin1", to = "UTF-8")

PolMine

# The get_sentiws function will download the zip-file with the SentiWS dictionary,
# unzip it and return a data.table.

library(data.table)

get_sentiws <- function(){
  
  sentiws_tmp_dir <- file.path(tempdir(), "sentiws")
  if (!file.exists(sentiws_tmp_dir)) dir.create(sentiws_tmp_dir)
  sentiws_zipfile <- file.path(sentiws_tmp_dir, "SentiWS_v1.8c.zip")

PolMine

After creating the Linux virtual machine on amazon EC2, the hard part was to set the Security Groups correctly to have port 8787 open, and to connect the newly defined Security Group to the instance. After that, everything ran smoothly as follows:

## Add CRAN mirror to sources.list, including key

```{sh}
sudo sh -c 'echo "deb http://cran.rstudio.com/bin/linux/ubuntu xenial/„ >> /etc/apt/sources.list'
gpg --keyserver keyserver.ubuntu.com --recv-key E084DAB9
gpg -a --export E084DAB9 | sudo apt-key add -
```

PolMine

The recent stable version of the package (v0.7.2) is available at CRAN:
https://cran.r-project.org/web/packages/polmineR/index.html

This is also where the new vignette / documentation can be looked at:
https://cran.r-project.org/web/packages/polmineR/vignettes/vignette.html

The package can be installed with the conventional package installation mechanism (from R):
```{r}
install.packages("polmineR")
```

PolMine

Installing a packaged corpus from the PolMine repository
--------------------------------------------------------

As an experiment, I have put a corpus of plenary procotols ("PLPRBT") into a private repository I host at the PolMine server. This is how to get it: You will need the devtools package to get the latest development version of polmineR. On Windows, installing devtools may require that you have installed Rtools.

```{r}
install.packages("devtools")
```
Now, install the development version of the polmineR package.