This project demonstrates a complete end-to-end sentiment analysis pipeline using Natural Language Processing (NLP) and a neural network built with TensorFlow/Keras. It processes and classifies customer reviews from Amazon, Yelp, and IMDB datasets to predict positive or negative sentiment.

This project uses ARIMA modeling to forecast telecom revenue based on two years of daily revenue data. It demonstrates how to prepare, analyze, and model time series data using both manual and auto ARIMA techniques, enabling short-term revenue forecasting to support strategic business decisions.