This dataset makes use of the “A Million News Headlines” dataset available at kaggle.com/therohk/million-headlines.

As a start, lets retrieve the dataset using pins and load is using readr:

library(pins)

news <- pin_get("therohk/million-headlines", board = "kaggle") %>%
  readr::read_csv()
## Parsed with column specification:
## cols(
##   publish_date = col_double(),
##   headline_text = col_character()
## )
news

First, lets clean up this dataset. For instance, the ‘publish_date’ column is a character, not a date:

library(dplyr)
## Warning: package 'dplyr' was built under R version 3.5.2
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
news_cleaned <- news %>%
  mutate(publish_date = as.Date(as.character(publish_date), format = "%Y%m%d"))

news_cleaned

Lets also assume we are interested in understanding when news outlets publish news, not what headlines are published. For this, we can keep the counts per day and month and throw away the other columns:

news_totals <- news_cleaned %>%
  group_by(publish_date) %>%
  summarize(count = n())

We can now plot by month the total amount of news being produced by this news outlet:

library(ggplot2)
## Warning: package 'ggplot2' was built under R version 3.5.2
news_totals %>%
  group_by(publish_date = lubridate::floor_date(publish_date, "year")) %>%
  summarize(count = sum(count)) %>%
  ggplot(aes(x=publish_date, y=count)) +
  geom_line()

Now, while this might complete our particular analysis, others in your team might be interested in easily fetching your tidied data set, which you can now easily share with pins in any available board. The following example uses RStudio Connect but this board can also be replaced to use Kaggle, GitHub or even custom boards:

pins::pin(news_totals, board = "rsconnect")

A colleague can now reuse your tidy dataset by fetching it from the source location:

pin_get("news-totals", board = "rsconnect")