---
title: "Setup test"
author: "Kenneth Benoit"
date: "23 April 2017"
output: html_document
---

```{r, echo = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE
)
```

#### Preliminaries: Installation

First, you need to have **quanteda** installed.  You can do this from inside RStudio, from the Tools...Install Packages menu, or simply using
```{r, eval = FALSE}
install.packages("quanteda")
```

(Optional) You can install some additional corpus data from **quantedaData** using

```{r, eval=FALSE}
## the devtools package is required to install quanteda from Github
devtools::install_github("quanteda/quanteda.corpora")
```

If you are feeling adventurous, you can install the latest build of **quanteda** from its [GitHub code page](https://github.com/kbenoit/quanteda).

Note that on **Windows platforms**, it is also recommended that you install the [RTools suite](https://cran.r-project.org/bin/windows/Rtools/), and for **OS X**, that you install [XCode](https://itunes.apple.com/gb/app/xcode/id497799835?mt=12) from the App Store.


#### Test your setup

Run the rest of this file to test your setup.  You must have quanteda installed in order for this next step to succeed.
```{r}
require(quanteda)
```

Now summarize some texts in the Irish 2010 budget speech corpus:
```{r}
summary(data_corpus_irishbudget2010)
```

Create a document-feature matrix from this corpus, removing stop words:
```{r}
ieDfm <- dfm(data_corpus_irishbudget2010, remove = c(stopwords("english"), "will"), 
             stem = TRUE)
```

Look at the top occuring features:
```{r}
topfeatures(ieDfm)
```

Make a word cloud:
```{r, fig.width=8, fig.height=8}
textplot_wordcloud(ieDfm, min.freq=25, random.order=FALSE)
```

If you got this far, congratulations!