The goal for this project was to develop insights into the contributing factors for both positive and negative customer reviews of a product or service. Analysing 60,000 customer reviews of airlines with a score between 1-10 on amenities such as seat comfort, cabin service, entertainment and value for money.

In this project I am using data analysis techniques to discover if certain amenity scores contribute to a negative or positive customer experience. This project implements and compares sentiment analysis techniques such as logistic regression, support vector machines and a recurring neural network on text documents. It also attempts to predict a customer's overall score based on the sentiment analysis performed on their written review of the airline and flight.

Target audiences for this project include customer and merchant focused organisations. The main technology used for this project will be Jupyter Notebook with Python. By using a notebook, the results of the code as well as some explanation in markdown text can be performed. For machine learning, the main model will be the use of TF-IDF and a Support Vector Machine (SVM).