2019 / TREND & BUSINESS RESEARCH
Fashion
Forward
From retail data to questions about fashion decisions.
Read the project ↓THE ORIGINAL STUDY / MARKETING MINOR
What can product data
tell us about a market?
A team study of Amazon retail information examined the relationships between price, product category, brand and sales using multivariate regression. Scatter plots and category comparisons were used to present the analysis.
TWO DISTINCT STAGES
01Original research
Product information, statistical modeling and visual analysis formed the course project. The archived description does not include the original dataset, coefficients or validation results.
02Later AI extension
A later proposal expands the questions to review sentiment, visual style classification and time-based trend forecasting. These are proposed extensions, not outcomes of the 2019 project.
03Questions for development
How could changes in customer reviews, product attributes and prices be studied over time? What additional evidence would be needed before forecasting trends or explaining sales changes?
LATER PROPOSAL / FROM DATA TO DECISIONS
01Collect & prepare
Proposed weekly retail collection, cleaning and normalization of products, prices, reviews and sales information.
02Read the signals
Review sentiment, visual style, price, brand and material attributes would become separate analysis inputs.
03Model & report
Explore sales models and time-based trends, then communicate findings through a dashboard and recurring reports. These remain proposed work.
PROJECT BOUNDARY
A study, followed by
a new proposal.
The early market study and the later AI plan remain separate in this archive. No prediction accuracy, business impact or running data pipeline is claimed.
Continue to the multiagent system proposal ↗