Headlines_train: Headlines_train

In mdsr: Complement to 'Modern Data Science with R'

Headlines_train

Description

This data comes from Chakraborty et. al., which combines headlines from a variety of news and clickbait sources. Some headlines contain subject matter inappropriate for classroom use. Given the volume of headlines containing such language (especially for clickbait == TRUE), this filtering might not catch all problematic headlines. User discretion is advised. The training dataset is a random sample of approximately 80% of the observations from the original dataset.

The testing dataset is a random sample of the remaining 20% of the observations not found in the training set.

Usage

Headlines_train

Headlines_test

Format

A data frame with 18,360 rows and 3 variables:

title

a character vector

clickbait

a logical vector

ids

an integer vector

A data frame with 4,589 rows and 3 variables:

Source

https://github.com/bhargaviparanjape/clickbait/

References

\Sexpr[results=rd]{tools:::Rd_expr_doi("10.1109/ASONAM.2016.7752207")}

Examples

nrow(Headlines_train)
nrow(Headlines_test)

mdsr documentation built on Sept. 11, 2024, 5:07 p.m.