Another Twitter Sentiment Analysis with Python - Part 4

Attached Jupyter Notebook is the part 4 of the Twitter Sentiment Analysis project I implemented as a capstone project for General Assembly's Data Science Immersive course.

Accompanying blog posts can be found from my Medium account: https://medium.com/@rickykim78

Below implementations can be found in the attached notebook.

Train/Dev/Test Split

Baseline

• null accuracy
• TextBlob sentiment polarity score (prerequisite: TextBlob pip install textblob)

Count Vectorizer

experiment with

• stop words removal
• varying vocabulary size
• n-grams