The dataset is a csv file containing Customers reviews about various Tunisian products and services.

We extracted more than 70 000 reviews collected from popular Tunisian applications on Google play store.

Your task is to classify the Customers reviews using NLP techniques in order to automatically understand if major users are satisfied or not.

Let’s get it started!

Variable definition:

• ID : Unique identifier of the text
• Content : Text_review
• score: Sentiment of the review (-1 for negative, 0 for neutral, 1 for positive)

Files available for download are:

• Train.csv - contains text on which to train your model.
• Test.csv - contains text on which you must classify using your trained model.
• SampleSubmission.csv - is an example of what your submission file should look like. PS: the names of the ID must be correct. Values in the 'score' column should -1, 0 or 1.