Skip to main content

Accessibility menu

Skip to main content Skip to footer
Return to calendar

Capstone Presentations

A Web Application for Sales Forecasting

Abstract:

 

The web application is developed for the sales forecasting of Walmart stores.  The Walmart Recruiting-Store Sales Forecasting dataset found on Kaggle is the data set used for project training and testing.  The dataset covers the period from February 5, 2010, to November 1,2012, featuring 45 Walmart stores from different regions and departments.  The key data fields include store numbers, department numbers, date, weekly sales, and five markdowns.

 

The web application gives users the option to select certain stores and departments to see weekly sales forecasts.  It features visualizations of past sales patterns, including year-over-year comparisons, to highlight peak sales seasons, trends, and anomalies detected by the model.  Besides, the system has performance metrics such as RMSE that instill user trust and confidence in the predications it displays.  The prediction model used is XGBoost, a model chosen by comparing different models with better prediction results.

When

  • 1:30 to 3:30 p.m. Monday, May 12

Where

Distance Education Rm, 102 Wing Technology Center

UWL campus map for building location and nearby parking lots.

University of Wisconsin - La Crosse Wing Technology Center

Contact

For questions about this event or to request disability accommodations , contact Sallie Coron at 608.785.6805 or scoron@uwlax.edu.

Parking

Payment may be required. No permit?
Use Passport Parking.

Additional parking info
Website for Parking | Email for Parking | Call for Parking

Loading...