Loading data
Last updated on 2025-07-22 | Edit this page
Estimated time: 0 minutes
Overview
Questions
- Which packages are needed?
- How is the dataset loaded?
- How is a dataset inspected?
Objectives
- Knowledge of the relevant packages
- Ability to to load the dataset
- Ability to inspect the dataset
Getting startet
When performing text analysis in R, the built-in functions in R are
not sufficient. It is therefore necessary to install some additional
packages. In this course we will be using the packages
tidyverse, tidytext and tm.
R
install.packages("tidyverse")
install.packages("tidytext")
install.packages("tm")
library(tidyverse)
library(tidytext)
library(tm)
Getting data
Begin by downloading the dataset called articles.csv.
Place the downloaded file in the data/ folder. You can do this directly
from R by copying and pasting this into your terminal. (The terminal is
the tab to the right of the console.)
R
download.file("https://raw.githubusercontent.com/KUBDatalab/R-textmining_new/main/episodes/data/articles.csv", "data/articles.csv", mode = "wb")
After downloading the data you need to load the data in R’s memory by
using the function read_csv().
R
articles <- read_csv("data/articles.csv", na = c("NA", "NULL", ""))
Data description
The dataset contains newspaper articles from the Guardian newspaper. The harvested articles were published on the first inauguration day of each of the two presidents. Inclusion criteria were that the articles contained the name of the relevant president, the word “inauguration” and a publication date similar to the inauguration date.
The original dataset contained lots of variables considered irrelevant within the parameters of this course. The following variables were kept:
- id - a unique number identifying each article
- president - the president mentioned in the article
- text - the full text from the article
- web_publication_date - the date of publication
- pillar_name - the section in the newspaper
Taking a quick look at the data
The ‘tidyverse’-package has some functions that allow you to inspect the dataset. Below, you can see some of these functions and what they do.
R
head(articles)
OUTPUT
# A tibble: 6 × 5
id president text web_publication_date pillar_name
<dbl> <chr> <chr> <dttm> <chr>
1 1 obama "Obama inauguration: We will… 2009-01-20 19:16:38 News
2 2 obama "Obama from outer space Whet… 2009-01-20 22:00:00 Opinion
3 3 obama "Obama inauguration: today's… 2009-01-20 10:17:27 News
4 4 obama "Obama inauguration: Countdo… 2009-01-19 23:01:00 News
5 5 obama "Inaugural address of Presid… 2009-01-20 16:07:44 News
6 6 obama "Liveblogging the inaugurati… 2009-01-20 13:56:40 News R
tail(articles)
OUTPUT
# A tibble: 6 × 5
id president text web_publication_date pillar_name
<dbl> <chr> <chr> <dttm> <chr>
1 132 trump Buy, George? World's largest… 2017-01-20 15:53:41 News
2 133 trump Gove’s ‘snowflake’ tweet is … 2017-01-20 12:44:10 Opinion
3 134 trump Monet, Renoir and a £44.2m M… 2017-01-20 04:00:22 News
4 135 trump El Chapo is not a Robin Hood… 2017-01-20 17:09:54 News
5 136 trump They call it fun, but the di… 2017-01-20 16:19:50 Opinion
6 137 trump Totes annoying: words that s… 2017-01-20 12:00:06 News R
glimpse(articles)
OUTPUT
Rows: 137
Columns: 5
$ id <dbl> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15…
$ president <chr> "obama", "obama", "obama", "obama", "obama", "oba…
$ text <chr> "Obama inauguration: We will remake America, vows…
$ web_publication_date <dttm> 2009-01-20 19:16:38, 2009-01-20 22:00:00, 2009-0…
$ pillar_name <chr> "News", "Opinion", "News", "News", "News", "News"…R
names(articles)
OUTPUT
[1] "id" "president" "text"
[4] "web_publication_date" "pillar_name" R
dim(articles)
OUTPUT
[1] 137 5Key Points
- Packages must be installed and loaded
- The dataset needs to be loaded
- The dataset can be inspected by means of different functions