Please don't post your solutions before the competition is over.

###File descriptions

- training_data.csv - 240,000 unique tweets about Covid-19 coming from six different English-speaking countries (each country has 40,000 tweets in total). Categorical outcome variable is called 'country'.

- test_data.csv - 60,000 unique tweets for which you will be predicting the 'country' variable. 'Id' variable shows the index values associated with each tweet.

- sample_submission.csv - a .csv file that shows the format using which you will submit your solutions for the competition ('Id' variable corresponds to the tweets found in the test dataset).

###Data fields

- text: The text of tweet (including emojis, htmls, hashtags)
- reply_to_screen_name: The Twitter screen name of the user the owner of the tweet is replying to (if any)
- is_quote: A Boolean variable that indicates if the owner of the tweet is quoting someone else’s tweet
- is_retweet: A Boolean variable that indicates if the owner of the tweet is retweeting someone else’s tweet
- hashtags: A list of hashtags included in the tweet
- country: The label of the country in which the tweet was posted (found only in the training dataset) 
- Id: An index number associated with tweets (found only on the test dataset)