The project was aimed to build scalable data warehouse in which the data are extracted by swarm drones and static roadside cameras. The ELT framework along with airflow helps analytic engineers in the city traffic department setup transformation workflows on the required objective. Airflow for scheduling tasks and DBT for transformation and build data model for analytics purpose was applied.
In this project, I have made analysis on the effectiveness of the ads using A/B hypothesis testing. The workflow involves reading the raw CSV data, preprocessing, identifying the control and exposed groups and presenting the result. I have made analysis using the classical and sequential A/B testing to measure the effectiveness of the ads. Additionally ML based A/B testing pipeline has been built.
In this project I have done data visualization that involves data preprocessing to reporting insights. The process involves reading raw CSV data, preprocessing, visualizations, polishing and reporting findings. Furthermore, more than 15 visualizations and five polished reports has presented. Finally slide deck are prepared to convey and reveal key summary of the insight made.
Pharmaceutical Sales Prediction across multiple Stores
In this project, I made prediction of daily sales in various stores up to 6 weeks ahead of time. The process involves preprocessing, data exploration and building predictive model with Machine and Deep Learning techniques. As a result I found the RandomForestRegretion algorithm performing better prediction.