EcommAI: Empowering Online Shopping with Intelligent Tools

Team Ashish N Mhatre & Ashwin Kadam

Problem Statement and Solution

Company ABC is struggling to manage their inventory stocks, predict future demand, and provide an enhanced customer journey on their website. With a small customer base, they need a solution that can provide insights to drive business decisions and improve customer satisfaction.

We, NEU-Analytics, propose the following solutions to address the challenges faced by Company ABC:

1. Build a demand forecasting model to predict future demand
2. Implement semantic search for products to ease customer journey from landing to checkout
3. Build a knowledge base based on YouTube review videos to perform generative QA
4. Build a product recommendation model to implement hybrid recommendations

Architecture:

1. Demand Forecasting

2. Recommendation System

3. QA System

Installation

To install the dependencies for this project, please run the following command:

pip install -r requirements.txt

Usage

To run the project, please follow these steps:

1. Clone the project repository
2. Run the Demand Forecasting notebook to (Setup snowflake environment, Transform and load data tables onto snowflake, Deploy Model, UDFs & Stored Proc)
3. Run Product_Associate notebook to (Setup pinecone environment, Transform data and train model to upload vector embeddings onto pinecone index)
4. Launch Ecommerce.py file using

Streamlit run Ecommerce.py

Project Report

EcommAI - Project Report

Streamlit App

EcommAI - Streamlit App

Video Demonstration

EcommAI - Video