Apriori Algorithm Implementation in Retail Data Mining

📌 Overview

This project implements association rule mining using the Apriori Algorithm, Brute-Force, and FP-Growth techniques. It provides a Flask-based web application for analyzing transactional data from major retailers (Amazon, Best Buy, K-Mart, Nike, and a generic dataset). Users can select datasets, set minimum support and confidence, and visualize the generated association rules.

🚀 Features

• Association Rule Mining using:
  • Brute-Force
  • Apriori Algorithm
  • FP-Growth Algorithm
• Flask-based Web Interface
• Dataset Selection
• Performance Comparison between algorithms
• Result Visualization with frequent itemsets and association rules

📊 Core Concepts

🔹 Frequent Itemset Discovery

Identifies items that appear together frequently in transactions.

🔹 Support & Confidence

• Support: Frequency of an itemset in the dataset.
• Confidence: Probability of item B appearing given item A.

🔹 Association Rules

Generated in the form: {A} → {B}, meaning if A is bought, B is likely to be bought.

🏗️ Project Workflow

1. Data Loading – Load transactional datasets.
2. Preprocessing – Encode data for analysis.
3. Parameter Selection – Users set support & confidence thresholds.
4. Algorithm Execution – Run Brute-Force, Apriori, or FP-Growth.
5. Result Comparison – Compare performance across algorithms.
6. Visualization – Display results in the web interface.

🛠️ Installation & Setup

🔹 Dependencies

Install required Python packages:

pip install Flask pyngrok mlxtend apriori_python