Skip to content

Marklong7/Dillard_return_forecasting

Repository files navigation

Return Forecasting

Description:

Utilize data science techniques with a PostgreSQL database and Python to analyze Dillard's dataset (10GB+), pinpointing the factors influencing customer returns. Employing algorithms like logistic regression, tree models, and Support Vector Machines (SVM), we are aiming to calculate the reduction of return rates, ultimately enhancing the profit of the business.

A meticulous Return on Investment (ROI) analysis showcases a robust ROI of 11.8%. Our strategic combination of tools and techniques promises to optimize Dillard's operations, curbing return rates and elevating the overall customer experience.

Key information:

Topic --- Analyzing and Classifying Customer Returns with Machine Learning Models.
Client --- Dillard's American department store chain.
Data --- 120 million records, 10GB+.
Business Question --- Predict product returns based on product information and transaction record.
Objective --- Optimize inventory management strategies maximize return on investment.

Team members:

  • Muhammet Ali Büyüknacar
  • Zhiwei Gu
  • Jialong (Mark) Li
  • Siyan Li