Skip to content

Latest commit

 

History

History
59 lines (33 loc) · 2.24 KB

README.md

File metadata and controls

59 lines (33 loc) · 2.24 KB

Machine Learning model to help retailers understand the properties of products and outlets that play crucial roles in increasing sales.

This project predicts the number of sales each food product will make using past data and provides insights on what features increase our sales.

Author:

Kamal Muhamed

Business problem:

Retailers cannot analyze their supply and demand and cannot forecast sales. They also don't have insights on what products get most sales and what stores generate the most revenue/sales. Our project aims to solve that for the retailers.

Data

The dataset used in this project is the sales_predictions.csv

Getting Started

Prerequisites

To run this project, you will need to have the following software installed on your system:

Python 3

Google Colab

Required packages - pandas, numpy, scikit-learn, seaborn, matplotlib

Results

Here are some of the insights after our analysis

bx_plt

corr

top5_pdcts

top3_stores

Model Used

I used the Regression model to make the sales predictions.

The R2 test score is 0.59

Limitations & Next Steps

Limitations

Limited feature set: This project only used a small number of features to predict sales, which may not be sufficient for all sales scenarios. There may be other features that are more predictive of sales but were not included in this project.

Next Steps

Incorporate more data: To improve the accuracy and applicability of the model, additional data could be collected from other regions or time periods. This could help to identify patterns and relationships that were not captured in the original dataset.

Acknowledgments

Scikit-learn for the machine learning library.

Matplotlib and Seaborn for data visualization.

For further information

For any additional questions, please email **Kamalmohapy@gmail.com