Getting Familiar with our Data

The Fake News data set from Kaggle contains over 7,500 rows or entries. I noticed some misaligned or improperly rendered rows, but generally this is our data size. As for fields there are 20 columns, one of which is ‘Type’ that┬ácategorizes the fake news entry. There seems to be some overlap among types, and “BS”, as the most common┬átype, is fairly ambiguous. However, you can get an idea of the types of fake news compiled.

Below are the total counts per Fake News Type:

 

Type Count
bias 443
bs 11444
conspiracy 430
fake 19
hate 245
junksci 101
satire 146
state 121

 

 

Advertisements