Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Retail Market Analysis

Ke Yuan, Yaoxin Liu, Shriyesh Chandra, Rishav Roy

TL;DR

The study tackles the challenge of understanding retail market dynamics by integrating heterogeneous datasets (Instacart orders, Walmart sales, Amazon reviews, Google Trends, and spending data) within a scalable Spark-based workflow. It systematically preprocesses and analyzes each dataset, then performs cross-dataset collaboration to reveal temporal, regional, and demographic patterns that drive demand and profitability. Key findings include seasonality and holiday effects on product trends, geographic and age-demographic drivers of sales, and the interplay between reviews, searches, and spending signals. The work demonstrates how big-data analytics can optimize inventory management, marketing strategies, and customer satisfaction, with practical pathways for future enhancements such as real-time analytics and expanded data sources.

Abstract

This project focuses on analyzing retail market trends using historical sales data, search trends, and customer reviews. By identifying the patterns and trending products, the analysis provides actionable insights for retailers to optimize inventory management and marketing strategies, ultimately enhancing customer satisfaction and maximizing revenue.

Retail Market Analysis

TL;DR

The study tackles the challenge of understanding retail market dynamics by integrating heterogeneous datasets (Instacart orders, Walmart sales, Amazon reviews, Google Trends, and spending data) within a scalable Spark-based workflow. It systematically preprocesses and analyzes each dataset, then performs cross-dataset collaboration to reveal temporal, regional, and demographic patterns that drive demand and profitability. Key findings include seasonality and holiday effects on product trends, geographic and age-demographic drivers of sales, and the interplay between reviews, searches, and spending signals. The work demonstrates how big-data analytics can optimize inventory management, marketing strategies, and customer satisfaction, with practical pathways for future enhancements such as real-time analytics and expanded data sources.

Abstract

This project focuses on analyzing retail market trends using historical sales data, search trends, and customer reviews. By identifying the patterns and trending products, the analysis provides actionable insights for retailers to optimize inventory management and marketing strategies, ultimately enhancing customer satisfaction and maximizing revenue.

Paper Structure

This paper contains 40 sections, 2 equations, 25 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Google Trends Data
  4. Feature Engineering
  5. Prediction
  6. Result
  7. Instacart Market Basket Analysis
  8. Data Profiling
  9. Data Pre-processing
  10. Market Basket Analysis
  11. Orders vs Hours of the Day
  12. Frequently ordered product departments and product aisles
  13. Relationship between position in shopping cart and reorder rate
  14. Percent Change in Consumer Spending
  15. Overview of the dataset
  16. ...and 25 more sections

Figures (25)

  • Figure 1: Design Diagram
  • Figure 2: Illustration of sin_season and cos_season
  • Figure 3: Distribution of Predicted Values
  • Figure 4: Denormalized Schema for Instacart Data
  • Figure 5: Orders vs Hours of the Day
  • ...and 20 more figures