page-header

Twitter US Airlines Sentiment Analysis

Project Description

We have all had bad airline experiences. However, some people take to Twitter (now X) to vent their frustrations. In this dataset, Twitter data was scraped in February 2015 of customers’ opinions of their airline.

I created a Natural Language Processing Model (NLP) in order to classify the tweets according to their sentiment. This helps us understand whether a customer is pleased with a certain airline or note.

I pre-processed the data then created two models, one with Count Vectorizer and the other with Tfidf Vectorisation. I then evaluated the results.

Photo credit: Kayak

  • This notebook is not available publicly due to course rules.

Client: AIML Program at University of Texas, delivered online by Great Learning

Skills: Natural Language Processing (NLP), Tokenization, Lemmatization, Count Vectorizer and Tfidf Vectorizer.