Sales Analysis & Demand Forecasting
Historical sales analysis and demand forecasting for a real Activewear Startup
- Streamed tweets into AWS S3 with Kinesis Firehose and combined it with a larger 55 mil.-tweet dataset (Not covered in this repo)
- Utilized PySpark in DataBricks to build custom PySpark transformers, label sentiment with SparkNLP/VADER, explore SparkML RandomForest and Logistic Regression classifiers, and to perform Latent Drichlet Allocation topic modelling
- Visualized results in AWS QuickSight through an Athena pipeline