Text Classification on Twitter Airline Sentiment Dataset from Kaggle.

Motivation:

• To familiarize with different ways to represent text for ML models(tf-idf, fasttext, embeddings).
• To familiarize with different ML libraries and framewords for NLP.
• To improve text data cleaning and processing skills.
• To explore text data augmentation techniques.

Content:

(Open the notebook with colab for better visualizations)

1. Exploratory Data Analysis on text data.

2. Text Data Preprocessing on Social Media Text Data, which is inherently noisy.

3. Modelling