Many consumers struggle with creating an efficient and comprehensive shopping list, often resulting in missed items during the shopping process. The aim is to develop a solution that recommends top 5 frequently bought items based on the selection of one item from a list, facilitating a more thorough and convenient shopping experience.
Problem Statement
Purpose
To create a user-friendly application that utilizes the Apriori algorithm for product recommendation and cross-selling, ultimately assisting users in building a well-rounded shopping list.
Project Goals
- Accepts a user's selection of a product from a list
- Applies the Apriori algorithm to recommend the top 5 frequently bought items associated with the selected product
- Provides the user with a selection mechanism among the recommended products
- Allows the user to add the chosen product to their shopping list
Technology Stack
Dataset: Online Retail dataset for comprehensive analysis
Python 3.10.10
Jupyter Notebooks
VS Code
Streamlit
Apriori Algorithm
Machine Learning
Methodology
1
Data Exploration
Investigated the Online Retail dataset to understand its structure and
relevant features
2
Algorithm Selection
Chose the Apriori algorithm for its effectiveness in mining frequent
itemsets
3
Data Preprocessing
Cleaned and preprocessed the data for input into the algorithm
4
Application Development
Used VS Code to create a Streamlit application that interacts with the
Apriori algorithm
Results & Screenshots
Landing Page - Product Selection Interface
Product Selection Process
AI-Powered Recommendations
User Selection from Recommendations
Shopping List Integration
Conclusion
The developed application successfully addresses the challenge of creating a comprehensive shopping list by employing the Apriori algorithm for effective product recommendations and cross-selling. Users can now enjoy a streamlined shopping experience, reducing the likelihood of forgetting essential items.
93% Accuracy
Real-time Recommendations
User-Friendly Interface
Scalable Solution