Skip to content

Latest commit

 

History

History
2 lines (2 loc) · 445 Bytes

File metadata and controls

2 lines (2 loc) · 445 Bytes

Twitter-Sentiment-and-Customer-Churn-Detection

Performed customer sentiment analysis on the tweets related to 4 major streaming services, applied various NLP and text analysis tools, and created multiple ensemble and deep learning classification models to implement churn detection and performed a rule-based feature extraction to find out the reasons for churn. Created a visualization dashboard using Plotly and Dash and deployed on Heroku.