data310

View the Project on GitHub

Final project Abstract

For my final project I am going to focus on how companies and brands can use machine learning to analyze the sentiment of Tweets. Twitter is a platform that allows users to engage with each other in real time, and stay up to date on the happenings of the world. Twitter also allows for consumers to voice their opinions on products and brands. The consumer data that is available on Twitter is essential to companies, but sometimes there is so much of it that it is hard for companies to sift through to find what is most relevant. This is where machine learning comes in. For my project I will focus on how companies can train a machine learning model to sift through Twitter data (Tweets) to analyze their sentiment. The goal of this project will be to create a model that companies can use to classify tweets regarding their brands. Hopefully the model will allow companies to feed in Tweets containing a keyword or phrase such as their brand name, and use indicators within the Tweet to do things such as sort the data into positive and negative product sentiment. The data I will use to do this project is within a python library called Tweepy. This will make the data gathering part of the project much easier. Additionally, because Tweepy exists, this also means that there are a number of scripts online that I can refer to for help while creating my model. At this point I am confident with where I am in the project, and I look forward to getting into the data.