Analyze how travellers in February 2015 expressed their feelings on Twitter.
The goal is to classify whether the sentiment of the text(present in the form of a tweet by someone) is negative, neutral or positive.
The dataset is taken from Kaggle: https://www.kaggle.com/datasets/crowdflower/twitter-airline-sentiment
In this project, analysis is performed on the above dataset and a classification model is created for predicting the sentiment of tweet, whether it’s positive, negative or neutral.
The repository contains two notebooks
- Analysis.ipynb: It contains the analysis, visualisations and conclusions made by analysing the dataset
- Classification_Model.ipynb: It contains the classification models and the results achieved.
We have used NLTK Naive Bayes, SVC and Random Forests for the classification Model. The results and accuracies are also compared.