Indonesian News Title Dataset

This datasets contains more than 90.000 Indonesian News Title collected from detik.com, one of the biggest Indonesian news portal.

The motivation behind this dataset is to enrich the resource in Indonesian NLP environment. This dataset is also very suitable for beginners to start working on real world data!

How Data is Collected

The data collection method is simple: it's web scraped using some Python libraries, such as BeautifulSoup4

Dataset Structure

In the CSV file, you'll find this columns:

Field	Definition
Date	Date when the news is posted
URL	Link to the news in detik.com
Title	Title of the news
Category	The news category. Enum: ['finance', 'food', 'health', 'hot', 'inet', 'news', 'oto', 'sport', 'travel']

So, each title is already labelled to a category so you can exercise lots of data science task on this data: supervised, unsupervised, etc.

For sample of the data, you can access the sample.csv

Data Description

Here are some characteristics of the data:

Category Distribution

Length Distribution

Date Distribution

Word Frequencies

Authors

If you want to contact the author of this dataset, reach me through:

• Email: ibamibrahim0 [at] gmail