The dataset is taken from Kaggle containing a collection of tweets labelled with the sentiment(positive or negative) associated with it. The following tweets were cleaned, parsed and segmented; counted most frequent words and hashtags associated with each emotion.

Formally, given a training sample of tweets and labels, where label ‘1’ denotes the tweet is negative and label ‘0’ denotes the tweet is positive, our objective is to predict the labels on the test dataset.

• id : The id associated with the tweets in the given dataset.
• tweets : The tweets collected from various sources.
• label : A tweet with label ‘0’ denotes positive sentiment while a tweet with label ‘1’ denotes negative sentiment.