A Portfolio containing all of the data projects I have done during my Master's and a few individual projects that showcase my analytical abilities.
This was a group project completed during my Master's program for the Info Visuals and Dashboards course.
Key Points of the project:
- Gathered and analyzed data related to game events, geolocation and weather, spectator voting, and inventory for MLB stadiums.
- Developed a Tableau dashboard to optimize inventory management and reduce food waste in MLB stadiums.
- Analyzed game events, geolocation and weather, spectator voting, and inventory data using metrics such as event type, rarity, quantity sold, and unit price.
- The Tableau dashboard includes components for food and beverage voting preferences, game event rarity and discount, food item selection, inventory data, and revenue tracking.
- Cleaned the inventory dataset and filtered voting datasets based on location.
- Merged datasets and performed necessary transformations for visualization in Tableau.
- The dashboard allows vendors to track game events, determine popular food and beverage items, manage inventory levels, and track revenue.
- Addressed challenges such as finding relevant data, visualizing qualitative food data, integrating votes, weather, and location, and ensuring consistent color schemes across visuals.
- The project scope focuses on optimizing inventory, minimizing waste, and maximizing profits for food vendors in MLB stadiums.
- Limitations include subjectivity in food preferences, data availability and accuracy, and the location-specific nature of the dashboard.
- Files related to the project are uploaded in the project repository for reference and detailed explanation of the dashboard.
Dashboards Created on Tableau:
Project 2: Analysing Social Determinants of Health – Age, Income, Crime with Respect to Health Insurance
This was a group project completed during my Master's program for the Data Wrangling course.
Key Points of the project:
- Identified groups at risk based on SDOH to target for insurance coverage.
- Explored ways to provide health insurance for vulnerable groups and determine appropriate coverage.
- Utilized publicly available SDOH and crime datasets for analysis.
- Employed data wrangling, preprocessing, enrichment, and visualization using Python (Jupyter Notebook) and Excel.
- Validated data quality, performed regression analysis, and developed risk scores.
- Explored expansion of analysis to other SDOH factors and geographic regions.
- Referenced external materials for additional insights into SDOH and health insurance.
- Detailed information is available in the final report in the project repository.
This was a group project completed during my Master's program for the Data Mining and Machine Learning Course.
Key Points of the project:
- Explored the growth and challenges in the online retail industry.
- Identified the need for customer segmentation to gain a competitive edge.
- Employed data mining techniques for automated customer segmentation.
- Used hierarchical clustering and K-means clustering for grouping customers based on shared characteristics.
- Leveraged the Online Retail dataset for analysis, including attributes such as InvoiceNo, StockCode, Quantity, CustomerID, etc.
- Cleaned and transformed the data to prepare it for analysis.
- Implemented K-means clustering to identify customer segments for personalized marketing campaigns.
- Utilized hierarchical clustering to visualize distinct customer groups.
- Recommended K-means clustering as a better choice for customer segmentation in this case.
- Highlighted the importance of data-driven decision-making and the benefits of machine learning algorithms in marketing strategy development.
- Provided managerial recommendations based on the clustering analysis.
- Detailed steps and visualizations are in the final report in the project repository.